Download The Relational Self: An Interpersonal Social–Cognitive Theory

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Document related concepts

Self-referential encoding wikipedia, lookup

Shelley E. Taylor wikipedia, lookup

Philosophy of experience wikipedia, lookup

Relational aggression wikipedia, lookup

George Kelly (psychologist) wikipedia, lookup

False consensus effect wikipedia, lookup

George Herbert Mead wikipedia, lookup

Social dilemma wikipedia, lookup

Group dynamics wikipedia, lookup

Impression formation wikipedia, lookup

Communication in small groups wikipedia, lookup

Personal identity wikipedia, lookup

Personalism wikipedia, lookup

James M. Honeycutt wikipedia, lookup

Interpersonal relationship wikipedia, lookup

Albert Bandura wikipedia, lookup

Social tuning wikipedia, lookup

Personality psychology wikipedia, lookup

Social perception wikipedia, lookup

Self-categorization theory wikipedia, lookup

Psychology of self wikipedia, lookup

Transcript
Psychological Review
2002, Vol. 109, No. 4, 619 – 645
Copyright 2002 by the American Psychological Association, Inc.
0033-295X/02/$5.00 DOI: 10.1037//0033-295X.109.4.619
The Relational Self: An Interpersonal Social–Cognitive Theory
Susan M. Andersen
Serena Chen
New York University
University of California, Berkeley
The authors propose an interpersonal social– cognitive theory of the self and personality, the relational
self, in which knowledge about the self is linked with knowledge about significant others, and each
linkage embodies a self– other relationship. Mental representations of significant others are activated and
used in interpersonal encounters in the social– cognitive phenomenon of transference (S. M. Andersen &
N. S. Glassman, 1996), and this evokes the relational self. Variability in relational selves depends on
interpersonal contextual cues, whereas stability derives from the chronic accessibility of significant-other
representations. Relational selves function in if–then terms (W. Mischel & Y. Shoda, 1995), in which ifs
are situations triggering transference, and thens are relational selves. An individual’s repertoire of
relational selves is a source of interpersonal patterns involving affect, motivation, self-evaluation, and
self-regulation.
ers in one’s life—as distinct from how the self is related to social
entities like groups or other social categories (e.g., Brewer, 1991;
Deaux, 1993; Tajfel & Turner, 1979). A significant other is defined as any individual who is or has been deeply influential in
one’s life and in whom one is or once was emotionally invested.
Thus, significant others may include members of one’s family of
origin, such as parents and siblings, as well as people encountered
outside of family relations, either early on or later in life—as in
one’s “chosen family” and friends. Our concept of the relational
self assumes that each of these significant others is linked to the
self, with each linkage capturing relatively unique aspects of the
self one is in relation to this significant other. The self is thus
entangled, shaped in part by ties with significant others, whether
these individuals are present physically or only symbolically (e.g.,
Baldwin, Carrell, & Lopez, 1990). One’s sense of self, including
thoughts, feelings, motives, and self-regulatory strategies, may
thus vary as a function of relations with significant others. Although we acknowledge that many aspects of the self do not
implicate any such relations, we suggest that the relations linked to
the self carry great importance for each individual.
The theory we propose builds on an extensive program of
research on the social– cognitive model of transference (Andersen
& Glassman, 1996), which examines the processes by which past
assumptions and experiences in relationships with significant others manage to resurface in relations with new people. Although we
have discussed some implications of this work for the self elsewhere (Andersen, Reznik, & Chen, 1997; S. Chen & Andersen,
1999), here we extend this thinking by proposing a theory that
articulates how various manifestations of the self and, more
broadly, personality can emerge in interpersonal contexts when
transference is elicited. The theory adds to the growing literature
on interpersonal approaches to the self (e.g., Aron, Aron, Tudor, &
Nelson, 1991; Baldwin, 1992; Higgins, 1987; Markus & Cross,
1990) as well as to the literature on social– cognitive views of
personality (e.g., Mischel & Shoda, 1995).
The social– cognitive model of transference, which undergirds
our theory, defines the phenomenon as occurring when a perceiv-
The nature of the self has long perplexed and intrigued scholars
across a broad spectrum of academic disciplines. In psychology
alone, well over a century of inquiry has translated into a virtual
explosion of theory and research in recent decades, especially in
social psychology, all aiming to chart the contours of the self.
Sharing this aim, we propose an interpersonal social– cognitive
theory of the self that draws on theory and research in social
cognition, personality psychology, and clinical psychology. Our
central argument is that the self is relational— or even entangled—
with significant others and that this has implications for selfdefinition, self-evaluation, self-regulation, and, most broadly, for
personality functioning, expressed in relation to others. The theory
clearly subscribes to the long-standing view that the self is fundamentally interpersonal (e.g., James, 1890). Indeed, we maintain
that an individual’s overall repertoire of relational selves, stemming from all his or her relationships, is a major source of the
interpersonal patterns that the individual enacts and experiences in
the course of everyday interpersonal life—whether at work, at
play, or in therapy.
The proposed theory focuses on the ways in which the self is
related to specific other individuals—namely, the significant oth-
Susan M. Andersen, Department of Psychology, New York University;
Serena Chen, Department of Psychology, University of California, Berkeley.
The research reported in this article was supported by National Institute
of Mental Health Grant RO1-MH48789 to Susan M. Andersen. Special
thanks to Noah Glassman, Inga Reznik, and Michele Berk, who have made
invaluable contributions to the research reported in this article. Thanks are
also due to Kathy Berenson, Christina Carter, Regina Miranda, Tami
Edwards, and Adam Klinger for their input.
Correspondence concerning this article should be addressed to Susan M.
Andersen, Department of Psychology, New York University, 6 Washington Place, 4th Floor, New York, New York 10003, or to Serena Chen,
University of California, Berkeley, Department of Psychology, 3413 Tolman Hall, Berkeley, California 94720-1650. E-mail: [email protected]
.nyu.edu or [email protected]
619
620
ANDERSEN AND CHEN
er’s mental representation of a significant other is activated in an
encounter with a new person, leading the perceiver to interpret the
person in ways derived from the representation and also to respond
emotionally, motivationally, and behaviorally to the person in
ways that reflect the self– other relationship (Andersen & Glassman, 1996). Although significant-other representations are idiosyncratic in content and meaning, much research supports the view
that the phenomenon of transference occurs by means of generalizable social– cognitive processes—that is, by means of the activation and use of these representations.
In line with the literature on knowledge accessibility (see Higgins, 1996c), our research has also shown that both chronic and
transient sources of accessibility play a role in the activation of
significant-other representations (Andersen, Glassman, Chen, &
Cole, 1995). Chronic sources derive from the frequency with
which a construct has been activated in the past, with high frequency resulting in a construct’s chronic readiness to be activated,
or its chronic accessibility (e.g., Bargh, 1999; Bargh & Thein,
1985; Higgins & King, 1981; Higgins, King, & Mavin, 1982). By
contrast, transient accessibility stems from cues in the environment. For example, cues that perceivers are exposed to prior to
encountering a stimulus person are priming stimuli that temporarily increase the accessibility of the construct, thereby heightening its likelihood of activation and use (e.g., Higgins, Rholes, &
Jones, 1977; Srull & Wyer, 1979). Another transient source of
accessibility lies in attended-to cues in a stimulus person that
“match” stored knowledge, thereby heightening the accessibility of
that knowledge (Higgins, 1989b, 1996c; Higgins & Brendl, 1995;
see also Hardin & Rothman, 1997; E. R. Smith, 1990). The term
applicability is often used to refer to this transient source because
the matching stimulus cues apply, or are relevant to, stored knowledge (e.g., Higgins, 1996c).
Research has also shown that significant-other representations
are chronically accessible, which means they have a special readiness to be activated regardless of contextual cues, although the
presence of transient cues heightens this chronic readiness further
(Andersen et al., 1995). In our research, we have relied heavily on
cues in a stimulus person that match stored knowledge about a
significant other—to transiently activate significant-other representations— because we view such applicability-based cues as an
analog for cues that perceivers might encounter in ordinary, faceto-face interactions (S. Chen, Andersen, & Hinkley, 1999). Such
cues need not be attended to in a conscious sense but instead may
contribute to knowledge activation even if registered only beneath
the threshold of conscious awareness (Glassman & Andersen,
1999a). Hence, we assume that people need not consciously draw
analogies with known others for transference to occur.
Because significant others are likely to be of such profound
importance in people’s lives, representations of them should not
only be chronically accessible but also highly laden with affect,
which implies that these representations play a pervasive role in
shaping interpretations and in defining emotional responses in
many, if not most, social encounters. In fact, given the emotional
and motivational relevance of significant others, transference is
likely to be profoundly characterized by emotional content (e.g.,
Andersen, Glassman, & Gold, 1998). Yet at the same time, we do
not concur with notions that transference is exclusively emotional
or defensive or that it operates without cognitive mediation, nor do
we assume that it is limited to therapeutic settings or to the
pathological or “neurotic” (J. R. Greenberg & Mitchell, 1983).
Instead, our work indicates that transference occurs in everyday
social contexts and among “normal” populations, thus depathologizing the nearly century-old concept. In short, prior experiences
with significant others continuously shape a broad range of personal and interpersonal responses in daily life.
What is most central about our existing model of transference
for the theory we propose here is its assumption that there are
linkages in memory between significant-other representations and
the self that reflect knowledge about who the self is in relation to
each significant other. Such linkages imply that when a perceiver’s
significant-other representation is activated, this activation spreads
to aspects of the self that are associated in memory with the
particular significant other—that is, to the relevant self-withsignificant-other (Hinkley & Andersen, 1996). As a result, interpretive biases in social perception emerge and, moreover, the
perceiver’s experience of the self comes to reflect, in part, the self
he or she experiences with the significant other— even though the
significant other is not there. This shift in the self involves shifts
in affect, motivation, and behavior (Andersen, Reznik, & Manzella, 1996; Berk & Andersen, 2000).
A Theory of the Relational Self and Personality
Our current theory extends these earlier assumptions about the
self and transference by embedding them within a more elaborated
and precise conceptualization of the nature of self-knowledge and
self-regulation and by considering how elements of relational
selves are linked to personality and how they unfold across situations. In particular, unlike our earlier work, which focused on the
idiographic nature of significant-other representations and of self–
other linkages, here we incorporate normative, categorical elements of relational selves in the form of interpersonal roles and in
the form of standards. We also integrate traditional individual
differences in personality, which tend to be conceptualized normatively (i.e., categorically), that may also be pertinent to
significant-other relationships. Doing so paves the way for a more
integrative model of self and personality in which self and
significant-other knowledge, however idiosyncratic in content, is
associated with normative constructs that also operate prescriptively in the context of relationships with significant others. Our
theory also extends our earlier work by conceptualizing how
self-regulatory processes transpire, a matter central to most theories of self and personality (as well as to much of contemporary
social cognition), and, in this way, treads still closer to the substance of personality as traditionally defined. Finally, we also spell
out how our theory offers an interpersonal if–then model of personality, a concrete case of the broader if–then framework put
forth by the cognitive–affective system theory of personality,
which defines personality in terms of person–situation interactions
(Mischel & Shoda, 1995).
In essence, we propose that the various selves-with-significantothers one has stored in memory compose a set of possible relational selves—a system of knowledge that comes into play in the
context of transference, with particular aspects of self– other
knowledge brought to the fore as a function of the particular
significant-other representation that is activated in the context. We
argue, then, that an individual’s overall repertoire of relational
THE RELATIONAL SELF
selves is an influential source of his or her interpersonal patterns—
and hence of the self and personality.
In this article, we first articulate the major propositions of our
theory and then present evidence supporting the theory. Later, we
consider related bodies of work on the self that are of special
substantive relevance because they touch on similar themes from a
broader vantage point in the field and thus locate our theory in the
larger theoretical discourse and raise integrative questions. With
the assumptions of our social– cognitive model of transference as
a backdrop, we present our theory of the relational self and
personality in five broad, overlapping propositions. The first elaborates on earlier assumptions about the nature of significant others
and the relationships people have with them while bringing to bear
new ideas central to our model of the relational self. The second
addresses variability and stability in the self, establishing our
theory as one that simultaneously addresses transient and longstanding influences on the self. Third, we argue that relational
selves embody a wide variety of self– other knowledge, both
idiographic elements that are unique to the self– other relationship
as well as socially shared aspects reflecting normative roles (e.g.,
an authority figure in relation to a less experienced person) and
self-standards (e.g., ideals the significant other holds for the self).
Fourth, we make the claim that our view of the relational self and
transference implies a theory of personality—not only in its stand
on stability and variability in the self, but also in its recognition of
both idiographic and nomothetic elements and its emphasis on
basic human motivations and self-regulation. Finally, our fifth
proposition posits that the model is a concrete case of the viewpoint that personality is best explained in terms of if–then relations. In accord with the cognitive–affective system theory of
personality proposed by Mischel and Shoda (1995), particular
situations evoke— because of their association with significant
others—specific aspects of personality, in this case, the relational
self.
Figure 1.
621
First Proposition: Relational Selves Are a Product of the
Profound Importance of Significant Others
At the crux of our theory is the idea that, given the profound
importance of significant others in people’s lives, the self and
personality are shaped largely by experiences with significant
others. The importance of these others derives from their emotional and motivational relevance for the self (e.g., Andersen et al.,
1998; Higgins, 1987). They carry emotional–motivational significance because, at the very least, they serve a self-regulatory
function. Emotional outcomes and motivational orientations often
hinge on the expectations, standards, and responses of those closest to us (e.g., Andersen et al., 1998; Downey & Feldman, 1996;
Maccoby & Martin, 1983; Mikulincer, 1998). Research on selfdiscrepancy theory, for example, has clearly shown that, when
salient, people’s beliefs about the standards their significant others
hold for them—and about whether they are meeting or failing to
meet these standards— deeply influence people’s emotional lives
and motivational focus (Higgins, 1987, 1989a; Higgins, Loeb, &
Moretti, 1995).
The emotional–motivational relevance of significant others is a
primary basis for our assumption that linkages are likely to be
formed and maintained in memory between knowledge about the
self and knowledge about significant others. Figure 1 offers a
simple depiction of our view of self– other linkages. Of course, the
general notion that others play a role in the self is widely recognized by contemporary researchers (Aron & Aron, 1996; Baldwin,
1992; Baumeister & Leary, 1995; Gergen & Gergen, 1988; Greenwald & Pratkanis, 1984; Higgins, 1989a; Markus & Cross, 1990;
McAdams, 1980; Ogilvie & Ashmore, 1991), just as it was by
early self theorists (e.g., Cooley, 1902; James, 1890; Mead, 1934;
Rogers, 1951; Sullivan, 1940, 1953). It is also an idea that cuts
across psychological subdisciplines. Research on attachment theory, which assumes that mental models of the self are shaped
Linkages between the self and significant-other representations in memory.
622
ANDERSEN AND CHEN
largely by early experiences with caregivers (e.g., Ainsworth,
Blehar, Waters, & Wall, 1978; Bombar & Littig, 1996; Bowlby,
1969, 1973, 1980; Bretherton, 1985; Thompson, 1998), is a good
example of this. Once examined mainly in clinical and developmental work, the theory has now made deep inroads into social and
personality psychology (e.g., Bartholomew & Shaver, 1998; Brennan & Morris, 1997; Collins & Read, 1994; Griffin & Bartholomew, 1994; Hazan & Shaver, 1994; Mikulincer, 1995; for a
review, see Simpson & Rholes, 1998).1
Our view of significant others and the self is especially compatible with the relational-schema approach (Baldwin, 1992). Like
ours, that approach is rooted in basic social– cognitive principles
and assumes, as we do, that linkages are stored in memory between
self and significant-other representations, with each linkage embodying the typical patterns of relating with significant others. On
the other hand, the approaches have diverged in emphasis.
Whereas much of the research on relational schemas has focused
on generic definitions of relational patterns (e.g., Baldwin, Fehr,
Keedian, Seidel, & Thomson, 1993) and of self-aspects (e.g.,
Baldwin & Sinclair, 1996), we have emphasized the unique nature
of significant-other representations and have focused on idiosyncratic conceptions of the particular self one is in relation to each
significant other in one’s life (Hinkley & Andersen, 1996; see S.
Chen & Andersen, 1999) and of course on the phenomenon of
transference. And although we now extend our thinking to normative constructs, the idiographic significant-other representation
remains our bedrock, as does the transference phenomenon.
Beyond depicting self– other linkages, Figure 1 also reveals our
assumption that people have multiple significant others, even
though there are likely to be individual differences in both the
number and the quality of significant-other relationships as well as
in the level of “significance” or intimacy experienced in them.
People are unlikely to have limitless significant others, however;
the numbers are rarely huge. In any event, one’s significant others
are those who are deeply important to the person and have had an
impact on his or her life. By definition, of course, there may be
profound differences within and across individuals in the specific
content of significant-other representations, which suggests, in
turn, that there may also be considerable variability in the nature of
the self one experiences with different significant others—including how adaptive or maladaptive the relational self and relationship tend to be. Although we have yet to systematically examine
all of these assumptions, our research has verified that people not
only experience little difficulty when asked to name various significant others but also exhibit considerable variability in how they
characterize these individuals, the relationships with them, and
their affective experiences and interpersonal roles in these relationships (Andersen et al., 1996; Baum & Andersen, 1999; see also
Ashmore & Ogilvie, 1992; Bacon & Ashmore, 1985; Baldwin,
Keelan, Fehr, Enns, & Koh-Rangarajoo, 1996).
Our contention that people form multiple significant-other
bonds fits well with the notion that there exists a fundamental
human motivation for belonging and connection (Baumeister &
Leary, 1995). Moreover, one criterion that has been used to establish belonging as a basic need lies in the concept of satiation,
which refers to the idea that, once belongingness needs are satisfied, people should be less motivated to form social bonds. Thus,
forming relationships adheres to a principle of diminishing returns
such that establishing new bonds is beneficial up to a point, after
which quantity becomes secondary to quality. To the extent that
significant-other relationships are particularly effective in satisfying belongingness needs, we suggest that it is likely that satiation
rules would apply to the formation of such relationships. Even if
satiation were not to fully account for limits on the number of
significant-other relationships one has, and other factors such as
time constraints or geographic location were to play a larger role,
there are likely to be limits on belonging needs (see Baumeister &
Leary, 1995).
That people have a limited number of significant others does not
imply a diminished role for these individuals in people’s lives.
Quite to the contrary, evidence suggests that people know so much
more about significant others than about anyone else (e.g.,
Andersen & Cole, 1990), other than the self (e.g., Prentice, 1990),
that even trivial or subtle tidbits of information that are observed
directly (or inferred on the basis of what is observed) about a new
person—such as interpersonal carriage, habits, inner qualities, or
ways of thinking and reacting—may seem to match knowledge
stored about a significant other (for a related argument, see Markman & Gentner, in press). Such applicability heightens the ordinarily chronic activation readiness of significant-other representations, thereby further increasing the likelihood of transference.
Although perceivers process information about others in a
bottom-up fashion to some degree, which should result in the
activation of whatever stored constructs happen to be applicable,
chronically accessible constructs, such as significant-other representations, are very likely to be brought to bear even in the absence
of relevant, individuating cues (Andersen et al., 1995). In short,
regardless of the precise number of significant others one has, the
activation of significant-other representations—and associated relational selves—is likely to be quite prevalent.
In defining significant others, it is worth acknowledging the
distinction between significant others who are “chosen” from those
who are “given” in the way members of one’s family of origin are.
For chosen significant others, it is feasible that transference may
have influenced not only one’s initial perceptions of these individuals but also one’s attraction to and ultimate choice of them. If
one gravitates to people on the basis of their similarity to a
previously known significant other (a prospect we have not directly examined), this might result in similarity between early and
later developing significant-other representations. Nonetheless, we
assume that new significant-other representations should eventually become relatively freestanding, at least in the sense that one
knows that the new person is not literally one’s parent, sibling, ex,
or other significant other, and that numerous distinctive qualities
are also likely. Hence, it should be the case that new significantother relationships different from those with given (or nonselected)
significant others should also form. Further, even if some degree of
1
The notion that significant others may be “included” in the self (e.g.,
Aron, Aron, & Smollen, 1992)—and are thus part of the self—is a different
perspective on the very considerable closeness of the self and significant
others. Although that assumption and the model based on it have been
fruitful both theoretically and empirically, they do not highlight specificity
in the self– other relationship or the self one experiences when with the
significant other, which are of fundamental concern in our model. We
regard this sense of inclusion as a metric of closeness with the other (see
also Aron & Fraley, 1999) rather than indicating there are no characteristics
that distinguish the self from the other.
THE RELATIONAL SELF
similarity ultimately exists between selected and nonselected significant others, this should not preclude each significant-other
representation from being relatively unique, nor the associated
relationship and relational self. Moreover, each significant-other
representation should evolve over time in its own way, being
revised and updated with continued use and experience, as with
any other representation (Conway & Ross, 1984; Loftus, 1982;
Loftus & Greene, 1980; see also Wachtel, 1981), and such changes
should favor differentiation by role and by other relationshipdefining features (see also Showers & Kevlyn, 1999). At the same
time, we assume that even with changes over time in significantother representations, there should be a degree of continuity in
one’s significant-other representations, as stored in memory (e.g.,
Demo, 1992; Greenwald, 1980; Rosenberg, 1979; Strauman, 1996;
Wylie, 1961, 1974), and thus in one’s repertoire of relational
selves.
Finally, although our theory is focused on interpersonal aspects
of the self, we acknowledge, of course, that self-knowledge involves many other aspects and domains, such as one’s abilities,
values, and goals (e.g., Higgins, 1996b). Indeed, even self-aspects
that involve significant-other relationships may include knowledge
that extends beyond significant others, such as personal standards
regarding how others in general should be treated. Nonetheless, the
profound emotional–motivational importance of significant others,
and the chronic accessibility of significant-other representations,
both suggest a strong, baseline influence on day-to-day functioning. Moreover, when combined with the presence of transient,
applicability-based cues in a new person, significant-other representations are especially readily triggered. Although representations of anyone one knows or has known, whether significant or
not, may exert an influence on social perception in this way, the
depth, extensiveness, and detail of the knowledge represented
about significant others, as well as both the mundane and sublime
nature of what is known, should make significant-other representations particularly likely to be activated in response to new
people. Thus, although self-knowledge is not limited to one’s
selves with significant others, these relationships are likely to play
an especially influential role in the nature of the self.
Second Proposition: Relational Selves Emerge in the
Context of Transference
Like many social– cognitive theories of the self, our theory
assumes that self-knowledge is extensive and well-organized in
memory (e.g., Cantor & Kihlstrom, 1987; Linville & Carlston,
1994; Markus, 1977), even though its exact structure remains an
open question (e.g., Higgins & Bargh, 1987; Higgins, Van Hook,
& Dorfman, 1988; Niedenthal & Beike, 1997; C. Showers, 1992).
We assume as well that given the complex array of knowledge one
has about the self, one’s entire pool of self-knowledge cannot be
cognitively accessible at the same time. Rather, only a subset of
this pool is in working memory at any given moment (e.g., Linville
& Carlston, 1994; Niedenthal & Beike, 1997). The working selfconcept is the term often used to refer to this subset (e.g., Markus
& Wurf, 1987), and it is widely believed that it is the working
self-concept, rather than any sort of monolithic self-representation,
that guides cognition, affect, and behavior on an ongoing basis
(Higgins & Bargh, 1987; see also Cantor & Kihlstrom, 1987;
Mischel & Shoda, 1995). In our theory of the relational self and
623
personality, we focus on the particular relational self that is active
in transference.
The activation of relational selves in transference. If the self
varies as a function of the self-knowledge that is currently active
in working memory, what determines which self-aspects are activated? Much research indicates that, just as for knowledge activation in general, different aspects of self-knowledge are rendered
accessible in part as a function of cues in the immediate situation
(e.g., Banaji & Prentice, 1994; Linville & Carlston, 1994; Markus
& Wurf, 1987; McGuire, McGuire, & Cheever, 1986). In essence,
the self is constructed anew as a function of the current context (cf.
Read, Vanman, & Miller, 1997; E. R. Smith, 1996). For example,
cues in one’s workplace are likely to elicit the set of cognitive,
affective, motivational, and behavioral responses associated with
one’s “professional self,” whereas cues in a party setting elicit
knowledge reflecting one’s “partying self.”
Recent years have witnessed the emergence of numerous approaches to contextual variability in the self, with context conceptualized broadly in some approaches (e.g., Deaux, 1993; Mischel
& Shoda, 1995) and more narrowly in others—such as in terms of
particular motives (e.g., Cantor, Markus, Niedenthal, & Nurius,
1986; Higgins, 1989a, 1996b), self-esteem contingencies (Crocker
& Wolfe, 2001), social roles (e.g., Linville, 1985; Markus &
Nurius, 1986), or relationships (e.g., Baldwin, 1992; Ogilvie &
Ashmore, 1991). Variability in the self is also central to theory and
research on present, actual, possible, and future selves (e.g., Cantor
& Zirkel, 1990; Higgins, 1987, 1996a; Markus & Nurius, 1986;
Moretti & Higgins, 1990; C. Showers, 1992). Our theory focuses
on variability in the self that occurs in contexts as a function of
whether or not transference is triggered. We argue that when a
significant-other representation is activated in a given context,
associated self-with-significant-other knowledge is brought into
working memory—that is, the working self-concept becomes infused with knowledge reflecting the relevant relational self.
Transient and chronic influences on relational selves. If relational selves emerge by virtue of the triggering of transference,
then contextual variability in the self that reflects different selveswith-significant-others ought to emerge in varying contexts depending on the same social– cognitive principles that govern the
activation and use of significant-other representations. Thus, these
sources of accessibility should combine to determine whether a
given relational self is activated. In terms of transient sources in
the immediate context, we focus on cues emanating from a new
person that make knowledge about the significant other applicable,
no matter how subtly. As noted, we view such applicability-based
cues as an ecologically valid source of transient activation for
significant-other representations, because they can be likened to
meeting someone who bears some resemblance to a significant
other (S. Chen et al., 1999). Thus, just as noninterpersonal cues,
such as those in one’s private office setting, are likely to elicit a
particular subset of self-knowledge reflecting the self one is when
alone at work, interpersonal cues in a new person, such as the way
he or she listens, holds one’s gaze, or draws one out, or even his
or her smell, gestures, facial features, habits, or attitudes, can all
serve as applicability-based cues that contribute to the activation of
a relevant significant-other representation, along with the associated relational self.
Although the notion of the working self-concept emphasizes
contextual variability in the self, we argue that chronic, long-term
624
ANDERSEN AND CHEN
influences attributable to the chronic accessibility of significantother representations operate alongside transient influences on
relational selves. The chronic activation readiness of significantother representations serves as a steady source of accessibility for
the relational selves that are linked to them— because the activation of a significant-other representation spreads to associated
self– other knowledge. Overall, then, variability in relational selves
should emerge on the basis of transient sources of accessibility
that contribute to the activation and use of significant-other representations, whereas continuity should derive from the chronic
accessibility of these representations. In short, the working
self-concept is shaped in part by which, if any, significant-other
representation is activated in a given context, because the activation of such a representation brings into play the relevant
self-with-significant-other.
Third Proposition: Relational Selves Have Both
Idiographic and Socially Shared Elements
As indicated above, the social– cognitive model of transference
is grounded in the literature on knowledge accessibility (Bargh,
Bond, Lombardi, & Tota, 1986; Higgins, 1989b, 1990, 1996c;
Higgins & King, 1981; see also J. S. Bruner, 1957; Kelly, 1955;
Sedikides & Skowronski, 1990, 1991; Wyer & Srull, 1986). In this
theoretical context, we have assumed that each mental representation of a significant other designates a specific individual, in
contrast to representations that designate shared notions of a social
category, type, or group, such as “Asians” or “politicians” (e.g.,
Andersen & Klatzky, 1987; Brewer, 1988; Cantor & Mischel,
1979; S. T. Fiske & Taylor, 1991; Higgins & King, 1981).
Significant-other representations are thus n-of-one representations,
which are also referred to as exemplars (Linville & Fischer, 1993;
E. R. Smith & Zarate, 1992). Although these representations
contain generic knowledge— generalizations about the person—it
is the significant person per se rather than a generic label that
accounts for the associations among bits of this knowledge.
The distinction between exemplars and generic constructs is
supported by evidence showing that memory is far more specific
than would be expected if only generic knowledge were retained
(e.g., E. E. Smith et al., 1998; E. R. Smith & Zarate, 1990, 1992).
In addition, considerable research has shown that exemplars are
used in social perception, as are generic social constructs, and that
exemplar-based processing is readily distinguishable from
category-based processing (e.g., E. R. Smith, Stewart, & Buttram,
1992). Indeed, we and others have shown that the significant-other
representation is one type of exemplar that is also chronically
accessible—relative to generic social constructs (e.g., Andersen et
al., 1995; S. Chen et al., 1999; Karylowski, Konarzewski, &
Motes, 2000).2
Consistent with exemplar notions, our past work has emphasized the uniqueness of each significant other in one’s life and has
argued that knowledge about each significant other is linked in
memory with relatively unique self-aspects and relational patterns.
We have thus assumed that significant-other representations house
various forms of idiographic knowledge about the other—for
example, physical characteristics and personality attributes (e.g.,
Andersen & Cole, 1990; Prentice, 1990), interpersonal behaviors
(e.g., Andersen et al., 1998), and inner feelings and motivations
(e.g., Andersen et al., 1998; S. Chen, 2001; Johnson & Boyd,
1995). When an idiographic significant-other representation is
activated, then, the changes reflecting the relevant relational self
should be idiographic as well.
The present theory continues to recognize the idiographic elements of significant others and relational selves but goes further by
positing that when an idiographic significant-other representation
is activated, this in turn activates not only idiographic self-withsignificant-other knowledge but also generic, socially shared constructs, such as social categories or social identities, that are linked
to the significant other. This contention is buttressed by recent
research showing, for example, that gender categories are automatically activated when significant-other representations are
primed, which is evidence for a link in memory between
significant-other representations and generic social categories (Karylowski et al., 2000). Applied to our theory, such data suggest that
activation of a significant-other representation should spread to
related normative constructs in the context of transference.
The possible forms of generic knowledge that may be linked to
a significant-other representation and self-with-other relationship
should, of course, cover a wide range. One possible form is the
interpersonal roles that the self and significant other maintain in
the relationship (e.g., A. P. Fiske, 1992; see also Bugental, 2000).
If relational selves include knowledge about one’s role in relation
to a significant other, then this role relationship should be invoked
when transference occurs in the context of an encounter with a new
person. As a result, role-based expectations, along with any affective consequences linked to such expectations, should play out in
the interaction with the new other (Baum & Andersen, 1999). For
example, a younger man may remind an older woman of her
younger brother, and she may thus take on the “older sister” role
with him, likewise expecting him to take on the “younger brother”
role with her. In a different version of this example, she may view
the exact ways her brother relates to her as an exemplar of how all
other younger men will relate to her, a phenomenon that would
render idiographic the semantic category designating this role.
Similarly, if a significant other was one’s “teacher” or “romantic
partner,” the ways in which the significant other behaved might be
used as an exemplar of how all “teachers” or all “romantic partners” will “relate to me.”
Generic, socially shared elements of one’s relational selves
should also include beliefs about the standards that significant
others hold for oneself. For example, they may include beliefs
about a significant other’s wishes about whom one should ideally
be (ideals), or about whom one ought to be (oughts). In selfdiscrepancy theory (Higgins, 1987; Higgins, 1996b), although
self-standards held from the standpoint of significant others are
assessed idiographically, they are defined in normative, categorical
terms. Our theory proposes that such normative elements of self–
other relationships are activated in transference. Specifically, when
a significant-other representation is activated, associated selfstandards should come to the fore, eliciting negative affective
consequences to the extent that one experiences a sense of failing
2
Of course, a generic category label may still call to mind a specificperson exemplar (E. R. Smith, 1992, 1998), and a specific-person exemplar
may also call to mind a generic category, because these kinds of knowledge
are connected in memory (see Karylowski et al., 2000).
THE RELATIONAL SELF
to meet these standards—for example, of not being who a significant other wishes one to be or believes it is one’s duty to be.
Fourth Proposition: Relational Selves Provide a Basis for
an Interactionist Model of Personality
Our fourth proposition is that relational selves provide a basis
for an interactionist model of personality. The crux of this contention is the idea that an individual’s overall set of relational
selves, from which derive aspects of the self that are experienced
in transference, is an influential source of one’s interpersonal
patterns. Below we describe the precise ways our theory of the
relational self extend into the realm of personality.
A Person ⫻ Situation model of personality. Until fairly recently, relatively little research systematically examined personality as a function of the varying situations individuals encounter
(e.g., Mischel & Shoda, 1995). This is in part because the field has
been slow to identify and agree on a language with which to
conceptualize interactive processes in personality. On the other
hand, the social– cognitive language of knowledge accessibility is
increasingly widely regarded as providing one way to frame such
models (Higgins, 1990; Mischel, 1999). This is the language of our
theory.
The social– cognitive language of knowledge accessibility allows us to apply an interactionist framework to our theory that
addresses personality psychologists’ aim of identifying regularities
in the person, as well as social psychologists’ aim of identifying
situations that have an impact on people across the board. We
argue that regularities in the person lie in an individual’s chronically accessible constructs—in particular, in the chronic set of
significant-other representations and associated self-withsignificant-other knowledge he or she has available in memory.
This set of representations, by virtue of its high activation readiness, is a steady source of the person’s responses, as noted above.
Coupled with these sources of continuity, there is the impact of
interpersonal situations that may trigger a perceiver’s significantother representation, consciously or nonconsciously, thereby enhancing the likelihood of its use and thus of the emergence of the
constellation of responses that defines the relevant self-withsignificant-other. Transference brings into play different relational
selves depending on the situation and what it cues. In this regard,
our theory reflects an interactionist model of personality, in which
personality is a function of both the person and the situation (e.g.,
Carson, 1969; Endler, 1984; Magnussen, 1990; Mischel, 1968,
1973; Pervin & Lewis, 1978).
Personality: Idiographic and nomothetic. Nomothetic differences in where individuals stand on specific trait dimensions have
been the classic basis for understanding individual differences in
personality. Rather than adopting an exclusively nomothetic, traitbased approach, however, we conceptualize personality in both
idiographic and nomothetic terms. On the one hand, we argue that
the idiographic constructs individuals have available in memory
are what shape their subjective interpretations. They endow these
interpretations with meaning and give psychological significance
to persons and events, which guides responses (Higgins, 1990;
Mischel, 1973, 1990; Mischel & Shoda, 1995; Nisbett & Ross,
1980). Hence, we argue that these idiographic constructs are as
important a source of individual differences as are traditional,
nomothetically defined individual differences. In fact, interpretive
625
and behavioral correlates of trait-based individual differences may
well be linked to the content of idiographic constructs and idiographic self-knowledge (e.g., Higgins, 1987; see also Dodge &
Price, 1994). For these reasons and more, personality theorists
have long argued that an idiographic approach sensitive to the
nuances of people’s lives and meaning systems is essential to
effective theory (e.g., Allport, 1937; Kelly, 1955). We concur that
individual differences are best understood, at least in part, in
idiographic terms—as implied by our idiographic treatment of
the contents of significant-other representations and of the
applicability-based cues that best activate them— even though we
define the processes underlying transference as generalizable
across individuals. In short, we assume that idiographic differences
in significant-other knowledge and associated self-withsignificant-other knowledge are critical to understanding personality (e.g., Epstein, 1983; Pervin, 1985; Snyder & Cantor, 1998).
On the other hand, our theory also incorporates normative,
categorical elements of self, relationships, and personality, as
noted above. Categorical individual differences in personality purported to involve significant others should thus interface with the
processes proposed in our theory. For example, individual differences in normatively defined self-standards held from the standpoint of significant others (e.g., oughts), and in discrepancies
between actual self-views and these standards (Higgins, 1987),
should elicit the specific affects and self-regulatory processes
predicted by self-discrepancy theory when an idiographic
significant-other representation is activated.
Motivational underpinnings of personality. Most broad-based
theories of personality have assumed that particular motivations
are key to how the self and personality develop and function, and
any adequate model of personality must therefore address matters
of motivation. In our theory, we take the view that motives and
goals are stored in memory as mental constructs and, like any other
construct, can be activated and thereby shape cognition, affect, and
behavior in goal-derived ways (e.g., Bargh, 1990, 1997; Bargh &
Gollwitzer, 1994). Motivations involving significant others, then,
may be stored as part of relational selves, so that when transference
occurs, the motives are then pursued in relations with new others.
What motivations are likely to be linked to relational selves? A
small number of fundamental human motivations are assumed
across a wide range of theoretical and empirical perspectives, and
amid these, the motivation that is perhaps most regularly identified
as a basic human motivation is the need for human connection—
for relatedness, belonging, caring, tenderness, or attachment (see,
e.g., Adler, 1927/1957; Bakan, 1966; Batson, 1990; Bowlby, 1969;
Deci, 1995; Fairbairn, 1952; J. R. Greenberg & Mitchell, 1983;
Guisinger & Blatt, 1994; Helgeson, 1994; Horney, 1939, 1945;
McAdams, 1985, 1989; Rogers, 1951; Safran, 1990; Sullivan,
1940, 1953). This motivation is the focus of a growing body of
work showing that being connected with others (or not) has consequences for cognition, affect, and behavior (e.g., Baumeister &
Leary, 1995; Leary, Tambor, Terdal, & Downs, 1995; Markus &
Kitayama, 1991; E. R. Smith, Murphy, & Coats, 1999) and,
indeed, mental health. Contingencies for being approved of or
loved by significant others, which is another way to conceptualize
connection needs, are also widely assumed to have profound
consequences for affect and behavior (e.g., Ayduk et al., 2000;
Bandura, 1986; Crocker & Wolfe, 2001; Downey & Feldman,
1996; Higgins, 1989a, 1991).
626
ANDERSEN AND CHEN
The need for human connection lies at the heart of our theory.
In our view, it provides the primary impetus behind the self’s
entanglements with significant others. Consciously or unconsciously, it is the “why” behind the initiation and maintenance of
these bonds. Moreover, it is the very basis for the significance of
these individuals in people’s lives—the desire to love and be
loved, to care and be cared for, to trust and to be trusted, to respect
and to be respected.
Of course, various other fundamental motives operate alongside
needs for connection—for example, needs for autonomy or freedom, for competence or mastery, for meaning, and for felt security,
the latter of which is to be out of harm’s way, literally and
figuratively (for a fuller discussion, see Andersen et al., 1997; see
also, e.g., Abramson, Seligman, & Teasdale, 1978; Bakan, 1966;
Bandura, 1977, 1986, 1989; Baumeister, 1991; Becker, 1971,
1973; J. Bruner, 1990; Deci & Ryan, 1985, 1991; Dweck &
Leggett, 1988; Epstein, 1973; Frankl, 1959; Glendin, 1962;
Guidano & Liotti, 1983; Heidegger, 1962; Horney, 1939; JanoffBulman, 1992; Jung, 1933; Klinger, 1977; Park & Folkman, 1997;
Pennebaker, in press; Seligman, 1975; Silver & Wortman, 1980;
Sullivan, 1953; White, 1959). Presumably, these needs can at
times work at cross-purposes with each other and with the need for
connection. As such, profoundly conflicting feelings can come to
be associated with significant others if one perceives that satisfaction of needs is unlikely or problematic. For example, the need for
connection might be blocked with a significant other if one were to
express another need, such as the need to behave more autonomously in the relationship. On the other hand, in our view, the need
for connection is so primordial that when it is blocked, it may well
be sought through any alternative means available, including
through paradoxical means, such as through a more exclusive
focus on competencies and accomplishments—if tenderness and
love are in short supply. Although our research has yet to examine
this, some of our evidence speaks to trade-offs and the ways in
which needs for connection may promote or alternatively bump up
against needs for felt security, as reflected, for example, in selfenhancement processes (Andersen et al., 1996; Hinkley &
Andersen, 1996). Overall, it is of considerable relevance to personality that such motives are linked to significant others, because
this implies that motivations involving significant others and expectancies based on outcomes experienced with these others
should be activated in transference.
Our model also builds on the motivational underpinnings of the
interpersonal theory of personality developed by Harry Stack
Sullivan (1953). Sullivan viewed personality as “the relatively
enduring pattern of recurrent interpersonal situations [that] characterize a human life” (p. 110). Not only does his model articulate
how varying aspects of the self arise as a function of situations in
which transference occurs, but it also assumes that motivation, at
the heart of most theories of personality, involves two basic needs.
First, there is a need for satisfaction, which encompasses the urge
to express oneself, including one’s own perceptions and emotions,
and to develop one’s competencies, while also experiencing tenderness and connection with others. Second is the need for security, which involves the urge to feel safe and protected.
Needs for connection, competence, and security, as we conceive
them, can thus be found in Sullivan’s model. And, coupled with
real interpersonal experiences with significant others, these needs
form the basis of what is contained in the “personifications” of
significant others and of the self that he argued people ultimately
form, as well as the “dynamisms” that reflect self– other relationships (Sullivan, 1953). The motivational material contained within
personifications and dynamisms is thus what is “transferred” in
transference, although because of many substantive shifts away
from classical Freudian theory (e.g., the drive–structure model),
Sullivan termed the transference phenomenon parataxic distortion. Unlike our theory, his was not a cognitive one, but it did
assume, as we do, that personifications and dynamisms associated
with significant others can emerge with new people, both inside
and outside of psychotherapy.3 Ours is a mental-representational
model (see also Singer, 1988; Wachtel, 1981; Westen, 1988) that
includes assumptions about fundamental human motivations.
Self-regulatory aspects of personality. Fundamental human
motivations are relevant to our theory not only because people
seek to satisfy these needs with significant others, but also because
failing to do so should be experienced as distressing and should
tend to provoke self-regulation, which in itself is closely tied to
personality. A consideration of the role of self-regulation in transference thus offers a window on how different elements of our
theory function.
Self-regulation involves overriding one’s own responses or
modulating them on the basis of some experience of threat, such as
an unpleasant emotional state (Carver & Scheier, 1982, 1998;
Larsen, 2000; Taylor, 1991; Tice & Bratslavsky, 2000). Although
self-regulation is multifaceted in its content and the processes by
which it occurs (see Kruglanski, 1996; Mischel, Cantor, & Feldman, 1996), one way it can work is that emotions stemming from
threat can function as a signal that something is off-kilter and that
an adjustment is needed (e.g., Baumeister, 1997; Schwarz & Clore,
1983; Wyer, Clore, & Isbell, 1999). Of course, such adjustments
are not always made—for example, when the signal is insufficiently alarming or when one lacks the necessary cognitive resources or the motivation to do so (Erber & Erber, 2000). We
argue, however, that the emotional–motivational relevance of significant others makes self-regulatory processes likely in the context of such relationships and thus important in understanding how
the relational self functions. Indeed, as described earlier, the importance of significant others derives in large measure from the
self-regulatory function they have (or have had) for one’s own
emotional life—for example, for one’s disappointments, hopes,
and fears (e.g., Andersen & Glassman, 1996; Bowlby, 1969;
Higgins, 1987, 1996a, 1996b; Higgins, Bond, Klein, & Strauman,
1986; Higgins et al., 1995; Higgins, Roney, Crowe, & Hymes,
1994; Maccoby & Martin, 1983; Strauman & Higgins, 1987).
Because remaining connected and feeling secure or protected
from harm are essential motivations, and are especially pertinent in
relationships with significant others, these motivations should be
reflected in what is regulated and why in relation to significant
others. Moreover, if self-regulation is wrapped up with significant
3
In Sullivan’s model, “real” interactions and “real” learning are influential in personifications and dynamisms (Sullivan, 1940, 1953), which
means that his theory also eschews the relatively exclusive reign of
unconscious fantasy proposed by Freud (1912/1958; see J. R. Greenberg &
Mitchell, 1983; Klein, 1932), who assumed that the realities of one’s
upbringing and treatment at the hands of others matter little because people
are largely lost in their own unconscious fantasies.
THE RELATIONAL SELF
others, then self-regulatory responses typical with a significant
other should be evoked in transference. More specifically, we
propose that the kinds of self-regulatory strategies that arise in
transference should be a function of the nature of the experienced
threat—that is, if it is a threat to the self or a threat to the
significant other. Threats to the self should undermine security and
safety and should thus elicit self-protective responses, aimed at
recovering a sense of self-esteem, capability, and security. Such
self-protective self-regulation should occur in transference when,
for example, the activated significant-other representation evokes
negative aspects of the self experienced with this other, which
constitutes a threat to the self.
Where threats to the self compromise security needs, threats to
the significant other cast one’s need for connection with the
significant other in peril. When such threats arise, relationshipprotective responses should be elicited. An example of such
relationship-protective self-regulation can be seen in research
showing that, when negative events in a valued romantic relationship are made salient, people tend to soften the negative implications of these events, presumably as a means of feeling better about
their significant other and the relationship (e.g., Murray & Holmes,
1993, 1994). Applied to the context of transference, when negative
aspects of a valued significant other are encountered in a new
person, this should pose a threat to the significant other and the
relationship, thereby provoking relationship-protective selfregulatory responses aimed at restoring one’s sense of connection
with the other. Moreover, in such a positive transference, one
should also expect the new person to regard one positively in the
way one’s significant other does or has, and should desire to be
emotionally close to the new person accordingly (Andersen et al.,
1996), all while diminishing the person’s negatives. The desire to
love and be loved by significant others—and contingencies for
same—should thus be stored in memory along with these representations (see also Higgins, 1996b, 1996d) and activated as part
of one’s relational selves. Overall, we argue that self-protective
self-regulation focuses on addressing security needs, whereas
relationship-protective self-regulation is primarily in the service of
connection needs.
Self-regulatory processes may also emerge in more complex
forms in transference. We propose a few such forms in an intersection between our model and self-discrepancy theory (e.g., Higgins, 1987, 1996a). In that theory, ideal self-standards (ideals) are
associated with a self-regulatory system that is attuned to attaining
(and not losing) positive outcomes, whereas ought self-standards
(oughts) are associated with a self-regulatory system attuned to
avoiding (and preventing) negative outcomes (Higgins, 1996b).
Put differently, ideals involve wanting to satisfy nurturance needs,
whereas oughts involve seeking to satisfy security needs. The
self-standards that are part of our theory are those held from the
perspective of a significant other. We propose that when a
significant-other representation is activated in an interpersonal
encounter, the ideals and oughts associated with the significant
other should be accordingly activated, thereby launching the corresponding self-regulatory system and subtle shifts in how one
self-regulates in the encounter.
Finally, we assume that self-regulation is most likely to occur in
transference when it is or was also a well-practiced response in the
actual relationship with the significant other. In this sense, our
theory may lay the groundwork for a more systematic tracking of
627
various kinds of suffering and resilience that may occur in transference (e.g., Andersen & Berk, 1998; Andersen, Chen, &
Miranda, 2002). In turn, it may carry wide-ranging implications for
mental health not yet understood, while at the same time providing
the underpinnings for conceptualizing the self and personality in
interpersonal and social– cognitive terms. The fact that our work
may have implications for the long-standing clinical assumption
that human suffering may result from inappropriately superimposing maladaptive responses learned in previous relationships onto
new relations, which remains of great interest to clinicians
(Andersen & Berk, 1998; Andersen & Glassman, 1996), adds to its
relevance to personality theory. For now, however, we limit our
scope to transference as a phenomenon that does not necessarily
involve suffering or “neurosis” and that occurs quite generally in
everyday social perception, while noting as well that the interpretive, affective, motivational, and behavioral consequences that
occur in transference may speak to such suffering.4
Fifth Proposition: Relational Selves are Cognitive–
Affective Units in an If–Then Model of Personality
As a natural extension of the preceding propositions, our fifth
and final proposition is that our theory can be seen as a concrete
case of Mischel and Shoda’s (1995) cognitive–affective system
theory of personality (see also Mischel, 1968, 1973, 1990, 1999).
Like ours, their theory takes a social– cognitive, interactionist
approach to personality, with the central argument that personality
functioning is best explained in terms of if–then relations.
If–then relations as basic units of personality. Countering
long-standing assumptions about cross-situational consistency in
trait-based personality responding, the if–then approach defines
personality in terms of the different responses (i.e., thens) that an
individual exhibits in different classes of situations (i.e., ifs). It
assumes that each individual possesses an idiosyncratic constellation of if–then relations and that the individual’s overall pattern of
if–thens reflects his or her unique “personality signature” (Mischel
& Shoda, 1995). Defining personality in these terms allows for
variability in personality responding across different situations
while capturing stability at the level of an individual’s signature.
Said differently, continuity in personality is conceived of as stemming from the predictability of cross-situational variability. From
this perspective, variability in an individual’s responding across
different situations should not be averaged over or dismissed—that
is, treated as error. Varying responses across situations are what
constitutes the individual’s personality.
4
Another aspect of our conceptualization of self-regulation—not tested
directly here—is our assumption that the person is an active organism—an
agent—who proactively regulates, maintains, and otherwise modulates
preferred psychological states. This assumption ultimately extends beyond
the individual as merely actively processing information to the selfdetermination of one’s own responses in accord with intrinsic interests,
dreams, and plans (Deci & Ryan, 1985). Self-regulation that is intrinsically
motivated— or self-determined—should facilitate health and well-being,
whereas that which is extrinsically motivated or controlled should not
(Deci & Ryan, 2000). This should hold in transference, as it does in other
contexts, where, for example, extrinsic control by a significant other may
compromise needs for autonomy and freedom, even while one seeks to
satisfy needs for connection by maintaining the relationship.
628
ANDERSEN AND CHEN
A key element of this if–then model is the idea that situations,
or ifs, are subjectively rather than objectively defined. Thus, objective situations carry idiosyncratic meaning for the individual.
The model argues that situations trigger specific cognitive–
affective units—such as encodings, expectancies, feelings, and
goals—which are what then give rise to the particular responses
the individual exhibits in these situations (Mischel & Shoda,
1995). The cognitive–affective units that are activated in a given
objective situation reflect the unique psychological situation experienced by the individual (Mischel & Shoda, 1995; see also S.
Chen, 2001; Higgins, 1990). It is this unique psychological situation that mediates the if–then relation—that is, the relation between the objective situation and the response exhibited in it. In
short, the model recognizes idiographic differences in how people
make sense of different situations and thus respond in them (see
also Higgins, 1990; Higgins et al., 1982; Kelly, 1955; Mischel,
1973).5
A growing body of evidence supports this if–then conceptualization of personality. It is impressive that this evidence has
emerged from in situ research in which behavior has been observed in a wide range of naturalistic situations and assessed over
time (e.g., Shoda, Mischel, & Wright, 1989, 1994; Wright &
Mischel, 1987, 1988).
Our theory of the relational self as a case of an if–then model of
personality. Our theory of the relational self is readily conceptualized in if–then terms. We too argue that variability across
different situations is fundamental to personality. In our case, we
focus on the variability that occurs by virtue of the activation of
different relational selves in interpersonal situations that trigger
transference. Our theory also converges with the broader if–then
model in the idea that idiographic psychological situations, rather
than objectively defined ones, mediate if–then relations. From our
perspective, significant-other representations and associated relational self-knowledge are the cognitive–affective units that, when
activated, yield the idiographic psychological situations that form
the basis for the particular responses individuals exhibit in particular settings.
At the same time, our theory is unique in its specification of
interpersonal situations and of the activation and use of significantother representations as social– cognitive processes that predict
variability and stability in personality. Accordingly, it is also
unique in its specific focus on those aspects of personality that
reflect relational selves. The focus is on interpersonal elements of
personality, and the assumption is that significant-other representations, coupled with associated self– other knowledge, constitute a
kind of interpersonal substrate of personality. This substrate is
interpersonal because it reflects the different selves one has in
relation to significant others. Each relational self embodies the
cognitive, emotional, motivational, and behavioral responses that
emerge in relations with the relevant significant other. Thus, the ifs
in our more targeted if–then model are situations involving other
people who, by virtue of their resemblance to a significant other,
trigger transference, and the thens are the interpretive tendencies as
well as affective, motivational, and behavioral responses associated with the relevant relational self that come into play. That is,
in our model, the new person one encounters is the interpersonal
situation. Hence, although we know that situations can and do
directly evoke meanings and various personal responses—as entering one’s childhood home may evoke desires to escape the grip
of one’s family—we focus on cues in a new person that evoke
knowledge of a particular other person and relationship, with
implications for how the self is experienced.
Overall, we view our theory of the relational self as an important
new member of the family of perspectives and models that conceptualize personality in if–then terms. Although the research
testing our theory, which we present in subsequent sections, uses
in situ methods in the laboratory rather than those in the real-world
contexts that have been used to test the broader if–then model (see
Mischel & Shoda, 1995), it is carefully controlled, experimental
work in which we have assessed a wide range of responses as a
function of contextual cues that trigger transference.
Summary of the Theory
To summarize, we have laid out a theory of the relational self
and personality in several broad, overlapping propositions. Our
starting point is the profound relevance of significant others to
people’s emotional and motivational lives, which paves the way
for linkages between significant others and the self. For each
individual, the result is a set of relational selves, each embodying
the relatively unique self one experiences in relation to the given
significant other, including the typical relationship patterns between self and other. Because relational selves are linked to
significant-other representations, when such a representation is
activated, the relevant relational self is activated accordingly,
infusing the working self-concept with knowledge that reflects the
self in relation to the significant other. In short, relational selves
emerge in contexts that trigger transference. This implies that
variability and continuity in one’s relational selves depends on the
same principles that govern the activation and use of significantother representations—namely, transient and chronic accessibility.
Variability arises as a function of contextual cues that trigger
transference, whereas continuity arises from the chronicity of
significant-other representations and associated relational-self
knowledge.
Although significant-other representations are exemplars, and
prior work has emphasized their idiographic nature, we have
proposed that relational selves contain both idiographic and socially shared elements. Thus, when transference occurs, idiographic as well as normatively defined shifts in the self should
occur—all reflecting the relevant relational self. That such shifts in
the self occur as a function of transference provides the basis for
an interactionist model of personality. We have argued that relational selves emerge as a function of the person—an individual’s
idiographic set of significant-other representations and associated
relational selves—and the situation, contextual cues that render the
activation of a significant-other representation more likely.
Our model is also tied to the personality literature in several
other ways. Namely, it shares a kinship with theories emphasizing
both idiographic and nomothetic elements of personality; has
motivational underpinnings, as most broad-based theories of personality do; and incorporates self-regulatory processes, which are
5
Indeed, the model explicitly argues that objective situations are not
what determine behavior. Instead, behavior depends on the “acquired
meaning of situational features” (Mischel & Shoda, 1995, p. 252), which is
idiographically defined.
THE RELATIONAL SELF
basic to personality functioning. Finally, our model is located in
the broader context of social– cognitive, interactionist theories of
personality and is a more targeted example of an if–then model of
personality.
The Evidence
Having laid out the key propositions of our theory, we now
present a body of research which, in general terms, demonstrates
that when a significant-other representation is activated, the self in
relation to this other is set into motion. The activation of this
relational self, which is one among the individual’s overall set of
selves-with-significant-others, involves the emergence of the specific set of emotions, motivations, and self-regulatory processes
associated with the self– other relationship, as well as shifts in how
one experiences the self when with this other.
Paradigm and Evidence for the Social–Cognitive Model
of Transference
As a backdrop for the findings most pertinent to our theory, we
first briefly describe our paradigm and the basic evidence for
transference (Andersen & Glassman, 1996). A distinguishing feature of the paradigm is its use of idiographic methods (Allport,
1937; Kelly, 1955; Lamiell, 1981) in a nomothetic, experimental
design (Andersen & Glassman, 1996; see also Higgins, 1990;
Pelham, 1993). This paradigm enables us to examine nomothetic
processes—namely, the activation and use of significant-other
representations— using content meaningful to each individual’s
life.
The paradigm involves two sessions. In the pretest session, we
ask research participants to name and describe their significant
others by generating short, idiographic sentences about them. They
are also asked to identify adjectives that are irrelevant to their
significant others. Some of these idiographic materials are then
used as stimuli in an ostensibly unrelated experimental session
conducted several weeks later, in which the irrelevant adjectives
serve as filler items.
In the experiment, participants are exposed to descriptive sentences about a new person, some of which were derived from those
they generated earlier (along with fillers) or from those a yoked
participant had generated. We thus manipulate whether or not the
new person resembles, to some degree, the participant’s own
significant other. In the resemblance condition, the significantother-derived descriptors serve as applicability-based cues for the
transient activation of the corresponding significant-other representation. Hence, transient activation of the participant’s
significant-other representation occurs in this condition and not in
the no-resemblance condition. In the latter condition, each participant is yoked with a randomly selected participant in the resemblance condition and is exposed to descriptors of that person’s
significant other. No 2 control participants are yoked with the same
person, and because each yoked pair encounters the same exact
descriptors, the precise contents of the stimuli are perfectly controlled. Of course, participants are randomly assigned to each
resemblance condition, enabling causal conclusions to be drawn
from this manipulation.
Significant-other-derived inference and memory about the new
person. Most of the research on transference has used one or
both of two standard measures of the phenomenon. The first is the
629
extent to which perceivers base their inferences and memory about
the new person on their stored knowledge about the significant
other—that is, the extent to which they go beyond the information
given (J. S. Bruner, 1957) about the person. Research has shown
that participants in the resemblance condition report higher
recognition-memory confidence about having been exposed to
descriptors that had not actually been presented, but that do characterize their significant other, as compared with participants in the
no-resemblance condition. Evidence across numerous studies has
shown such inference and memory effects (Andersen et al., 1995,
1996; Andersen & Baum, 1994; Andersen & Cole, 1990; Baum &
Andersen, 1999; Berk & Andersen, 2000; S. Chen et al., 1999;
Glassman & Andersen, 1999a, 1999c; Hinkley & Andersen, 1996).
The data have also shown that such effects are reliably stronger for
significant-other representations than they are for nonsignificantother representations or for social categories such as stereotypes
(for a review, see S. Chen & Andersen, 1999). Hence, the phenomenon cannot be reduced to what occurs for any person representation or for any social category, nor to global implicit theories
about people. Research has also indicated that the effects cannot be
reduced to self-generation effects (Greenwald & Banaji, 1989).
Finally, research has shown that it occurs for both positively and
negatively regarded significant others and thus is not due to
affective valence (e.g., Andersen & Baum, 1994; Andersen et al.,
1996).
Finally, as noted in prior sections, research has demonstrated the
chronic accessibility of significant-other representations. They are
chronically ready to be used even in the absence of transient
priming (Andersen et al., 1995, Study 1) and when applicabilitybased cues are minimal (S. Chen et al., 1999) or absent (Andersen
et al., 1995, Study 2). Yet both transient priming and transient
applicability-based activation contribute to the phenomenon
(Andersen et al., 1995), which is robust enough to persist and even
to be exacerbated over time (Glassman & Andersen, 1999c).
Significant-other-derived evaluation of the new person. The
second standard measure of transference is the degree to which
participants base their evaluation of a new person on their evaluation of the significant other. Research has shown that people tend
to like a new person more when he or she resembles a positively
regarded significant other than a negatively regarded other—an
effect that has not occurred in the no-resemblance, control condition (Andersen et al., 1996; Andersen & Baum, 1994; Baum &
Andersen, 1999; Berk & Andersen, 2000; Reznik & Andersen,
2001). In accordance with the theory of schema-triggered affect
(S. T. Fiske & Pavelchak, 1986), the overall evaluation of the
significant other is applied to the new person. In that process, a
“summary” evaluation is linked to representations in memory, so
that a person classified and interpreted in terms of such a representation is regarded positively or negatively in a parallel manner.
Indeed, some evidence for comparable effects in self-reported
mood states exists (Andersen & Baum, 1994), although it does not
always replicate (Andersen et al., 1996). We interpret these mixed
results for mood as suggesting that affect (vs. evaluation) may
emerge through a more nuanced process than the evocation of
overall summary evaluations as evidenced in research described
below. Indeed, we assume that most significant-other representations are unlikely to be exclusively positive or negative, as people
may often have ambivalent feelings about their significant others.
Nonetheless, our research focuses primarily on contrasting positive
630
ANDERSEN AND CHEN
and negative significant others—simply to control variability in
overall affect associated with the representation. At the same time,
because we maintain that no significant other is likely to be
affectively “neutral,” given how laden with affect each is, we have
not attempted to construct a neutral significant-other control group.
Significant-other representations activated outside of awareness. Because effortlessness in the activation and use of
significant-other representations can speak to the potential ubiquity of the transference phenomenon, and nonconscious activation
can rule out the necessity of being consciously reminded of a
significant other as a precondition for transference, evidence for
the nonconscious triggering of transference is important (Glassman & Andersen, 1999a). The notion that transference occurs
unconsciously is also central to the clinical concept (e.g., Ehrenreich, 1989; Glassman & Andersen, 1999b; Luborsky & CritsChristoph, 1990). We assume that transference can be triggered
outside of conscious awareness and have demonstrated this using
a variant of our basic paradigm in which cues about the new person
were presented subliminally. In this research, participants sat at a
computer terminal for a “computer game” with a partner seated
elsewhere. As part of this game, they were told to focus their
attention on a white dot in the center of a computer screen in order
to read supraliminal stimuli presented above the dot, while simultaneously responding to random flashes on the left and right of the
screen by pressing “left” or “right” on a response box. These
flashes were actually subliminal descriptors, each consisting of
four or fewer words, and flashed for less than 100 ms in parafoveal
vision and then pattern masked (as in Bargh et al., 1986). The
supraliminal stimuli were always irrelevant to the participant’s
significant other. In contrast, the subliminal descriptors were derived from the participant’s own significant-other descriptors in
the resemblance condition and from a yoked participant’s in the
control condition.
After the game, participants completed an inference measure
about their partner. As predicted, participants in the resemblance
condition made stronger significant-other-derived inferences than
did no-resemblance participants— even though each pair of yoked
participants was subliminally exposed to the same descriptors
(Glassman & Andersen, 1999a). Finally, an additional control
condition ruled out self-generation effects (of the subliminally
presented stimuli), and a subliminality check verified the stimuli
were presented outside of awareness. The data thus show that
transference can be triggered nonconsciously and does not depend
on perceivers being consciously reminded of a significant other.
Transference Evokes Closeness Motivation, RelationshipProtective Self-Regulation, Affect, and Behavior
We now turn to two lines of research that speak directly to our
theory of the relational self and personality. In the first, we focus
on the assumption that significant-other representations are affectively and motivationally charged because of their inextricable
relation to the self. Hence, when a significant-other representation
designating someone to whom one feels (or once felt) close is
activated, motives and feelings experienced with this other should
emerge in relation to the new person. In particular, the basic
human motivation to be connected with the other should be set into
motion so that people are more likely to seek closeness with the
new person—to approach him or her—if he or she resembles a
positively regarded rather than a negatively regarded significant
other. Heightened approach tendencies under these circumstances
should also mesh well with higher expectations of acceptance
rather than rejection from the new person, which should also
emerge. Along with these motivationally infused elements, positive emotions associated with a positive significant other should be
experienced in transference, and self-regulatory processes should
be set into motion by any threat to these positive experiences in an
effort to fend off the threat. Finally, the significant-other-derived
emotional and motivational experiences should be profound
enough to influence dyadic interpersonal behavior in transference,
such that the relationship with the significant other is behaviorally
re-created with the new person.
In research examining these issues, participants in both the
resemblance and no-resemblance conditions were exposed to descriptors about a new person derived from a positively or negatively regarded significant other, their own in the resemblance
condition and a yoked participant’s in the no-resemblance condition. Participants were randomly assigned to one of these four
conditions. In all conditions, participants were exposed to equal
numbers of positive and negative descriptors regardless of the
overall positivity or negativity of the significant other so that
descriptor valence could be ruled out as an explanation for the
observed effects. In each study below, the data verified that transference occurred in terms of one or both of our standard measures
of transference—that is, inferences and memory about, and evaluation of, the new person.
Activating the basic need for human connection. We assume
that the affect-laden nature of significant-other and self-withsignificant-other knowledge should mean that these representations include motivational material. We have focused on the interpersonal motivation to be emotionally close with others because
this basic need is likely to have particular resonance in significantother relationships and hence should be stored in memory in the
linkages between the self and the significant other. Therefore,
when a representation of a positive versus a negative significant
other is activated in an encounter with a new person, the desire to
emotionally approach versus avoid the person, respectively, should
emerge because activation of this representation should in turn
elicit relevant relational-self knowledge. As predicted, research
has shown that participants were more motivated to be emotionally
open, and not distant, with a new person when the person resembled their own positively versus negatively toned significant other,
an effect not seen in the no-resemblance condition (Andersen et
al., 1996; Berk & Andersen, 2000; see also Reznik & Andersen,
2001). The need for connection is thus evoked in transference.
Activating interpersonal expectancies for acceptance or rejection. The need for connection with significant others makes it
likely that people’s bids for acceptance and love, and the others’
responses to these bids, will be retained in memory. For example,
if a significant other is harsh and rejecting, stored knowledge about
this significant other and the self– other relationship will surely
include expectancies for rejection and responses to this. Indeed,
contingencies for acceptance and rejection with significant others
may well be fundamental to significant-other relationships and to
our understanding of ourselves in relation to others (e.g., Bandura,
1986; Downey & Feldman, 1996; Higgins, 1989a, 1991). Our
theory assumes that stored linkages between the self and significant others include such interpersonal expectancies, which should
THE RELATIONAL SELF
play out in transference. As predicted, research has shown that
participants have higher expectations of being accepted by a new
person when the person resembles one of their own positive versus
negative significant others, a pattern not observed in the noresemblance condition (Andersen et al., 1996; see also Reznik &
Andersen, 2001). It is important to note that the self– other relationship in this case runs not from the self to the significant other,
as it does for the motivation to be close and connected with the
other, but rather runs from the significant other to the self—as it
reflects one’s perceptions of the other’s feelings— of acceptance or
rejection. Together, these data reveal the bidirectionality of the
relational dynamics between self and significant others as stored in
memory.
Activating facial expressions of affect. The fact that
significant-other representations are affectively laden suggests that
the emotional meaning of significant-other relationships should be
experienced in transference. We tapped this using a modality of
emotional expression that is widely studied—facial expression.
Specifically, we covertly videotaped participants’ immediate facial
movements while they read each descriptor presented about a new
person. Naive judges then rated the pleasantness of their facial
expressions (adapted from Tesser, Millar, & Moore, 1988). As
predicted, when the new person resembled the participants’ own
positive significant other, rather than a negative one, their facial
expressions revealed more pleasant affect; no such effect occurred
in the no-resemblance condition (Andersen et al., 1996). Hence, in
the resemblance condition, the overall affective tone of participants’ significant-other representations emerged in facial expression in response to each descriptor. This effect was observed
averaging across both positive and negative valence in the individual descriptors.
Activating relationship-protective self-regulation on the basis of
threat cues. As indicated, the motivation for emotional closeness
and connection is activated in the context of a positive transference, as are expectations of acceptance and positive affect. Encountering negative descriptors about a new person that happen to
be characteristic of a positively regarded significant other should
therefore pose a threat to the positivity with which one views the
significant other and the relationship, thereby precipitating a selfregulatory response. Specifically, when negative cues suggest that
the new person possesses disliked qualities of a positive significant
other, threat should ensue unless a compensatory process occurs
that somehow softens the negative implications of the threatening
information. Data from the research described above on facial
affect support this very prediction (Andersen et al., 1996). Namely,
in the resemblance conditions, participants responded to negative
descriptors about the new person that reflected disliked characteristics of their positive significant other by showing more pleasant
facial affect in response to these negative descriptors—relative to
any other condition of the study and, in particular, relative to
positive descriptors about this same significant other. Hence, the
overall positive tone of the significant-other representation drove
facial affect, such that participants’ previous evaluations of the
negative characteristics encountered in this condition were reversed in their facial affect, presumably as a self-regulatory response. No such effect emerged in the no-resemblance conditions.
In our view, this finding reflects the need for connection with
positive significant others—the desire to maintain a positive relationship with the significant other in spite of his or her flaws.
631
Being reminded (consciously or unconsciously) of negative attributes of a significant other in the context of transference may
otherwise pose a threat to one’s connection needs. Finding ways to
perceive these negative attributes positively may be critical in
significant-other relationships and may thus be well practiced.
Research on romantic relationships has shown that people often
work to soften or neutralize negative attributes of their romantic
partners and that recalling negative events in a romantic relationship may even result in a more positive evaluation of the partner,
presumably because the need to enhance the other emerges in the
face of threatening facts (e.g., Holmes & Rempel, 1989; Murray &
Holmes, 1993, 1994). The fact that positive transference experiences include experiencing a desire to be close and expectations of
acceptance, rather than the desire for emotional distance and
expectations for rejection as in the case of a negative transference,
further suggests that this self-regulatory response in transference
may serve to protect the relationship with the significant other.
Activating significant-other-derived interpersonal behavior.
Our theory assumes that the reemergence of affective qualities of
the relationship with a significant other in the context of transference is likely to be profound—and enough to influence actual
interpersonal behavior in a dyad. When a perceiver’s significantother representation is activated in a dyadic interaction with a new
person, the behavioral dynamics should come to reflect the affect
the individual associates with the relevant significant-other relationship. Specifically, because transference elicits positive or negative beliefs, feelings, and motives in line with the overall affect
associated with the significant other, it should set into motion a
cycle of behavioral confirmation (Berk & Andersen, 2000). In
such cycles, perceivers’ assumptions create a corresponding social
reality by eliciting behaviors from others that confirm what was
previously in the perceivers’ minds (e.g., Rosenthal, 1994; Snyder,
1992). Although there are limiting conditions on when and how
behavioral confirmation will occur, substantial evidence indicates
that it clearly does (Snyder, 1992).
In research examining behavioral confirmation in transference,
after participants were exposed to descriptors about a new person,
they engaged in an audiorecorded conversation with the person,
who was a total stranger and who was entirely naive about the
purpose of the study (adapted from Snyder, Tanke, & Berscheid,
1977; see also Andersen & Bem, 1981). Each stranger was randomly paired with each participant, and hence bore no relation to
the participant’s significant other, thus constituting a more conservative test than an in situ context in which a new person might
bear enough resemblance to the significant other to actually trigger
transference.
Conversational behavior was assessed, focusing on the pleasantness of affect expressed in the new person’s contributions to the
conversation. Independent judges rated an audiotape of these contributions, excluding the participant’s side of the conversation to
be able to assess the new person’s behavior uncontaminated by the
participant’s. Evidence for behavioral confirmation in transference
was found, in that the new person expressed more positive affect
in the conversation in the positive significant-other-resemblance
condition than in the negative resemblance condition. This effect
did not occur in the no-resemblance conditions. These data thus
demonstrate that behavior elicited from a new person in the context of transference can be predicted from the affective quality of
the relationship with the relevant significant other, with the new
632
ANDERSEN AND CHEN
person coming to respond behaviorally as the significant other
typically does (or did).
Summary. To summarize thus far, we argue that this first set of
findings speaks in multiple ways to our overriding argument that
self and personality can be traced in part to transference. When a
significant-other representation is activated in an encounter with a
new person, the affective, motivational, and behavioral elements
that characterize the self in relation to the relevant significant other
are set in motion. In the context of transference, desires for
closeness or distance, expectancies for acceptance or rejection,
facial affect, and interpersonal behavior in the encounter with the
new person are colored by the affective and motivational qualities
of the relevant self– other relationship. Moreover, the data show
that self-regulatory responses occur in transference to fend off
threats to the relationship, which would be expected if the affective
and motivational ties linking the self and significant others come
into play in the transference phenomenon.
Transference Evokes Self-Evaluation, Self-Protective
Self-Regulation, and Self-Discrepancies
The next body of evidence addresses still more precisely our
assumption that relational selves are activated and played out in
transference. We have argued that the activation of a significantother representation elicits shifts in the working self-concept toward the self with the relevant significant other. These shifts
should involve both idiographic and normative elements of the
relational self. As for idiographic shifts, the working self-concept
should come to be characterized, at least in part, by the idiographic
attributes one associates with the relational self. Moreover, one’s
evaluation of these changes in self-concept attributes should reflect
the overall affect associated with the relationship. That is, all other
things being equal, one comes to feel relatively good or bad about
the self as a function of how one feels toward the significant other.
Of course, as we describe later, it is possible to feel bad about the
self in relation to a person one loves, with the affect experienced
deriving from how the self is experienced and evaluated in the
relationship.
At the same time, self-regulatory processes should occur when
shifts in the working self-concept pose a threat to the self, such as
when these shifts involve an influx of negative self-attributes. In
this context, compensatory self-enhancement should occur— calling to the fore, as a self-protective response to a threat to the self,
competing, positive aspects of the self. Indeed, these processes
may be part of the common experience of, for example, many
college students who go home to be with family at the holidays
only to find themselves being young kids again in ways that are
quite unpleasant, and thus use other more recently acquired aspects
of self to try to bolster the self against fully succumbing to this
unpleasant experience.
As for normative shifts in the working self-concept, we argue
that the activation of a significant-other representation should in
turn activate normative social constructs that characterize the self
with the significant other, such as the respective interpersonal roles
occupied by the self and the other, as well as the particular
normative standards the other holds for the self. As a result, the
affect associated with such normative self-with-other knowledge
should arise in transference. For example, if the normative knowledge involves interpersonal roles, then roles should be activated in
transference, and the affective consequences assumed to occur
when roles are violated should emerge if a role violation were to
occur in this context. Or, if the normative knowledge involves
standards that a significant other holds for oneself and one fails to
meet these standards, the distinct affective consequences of activating particular standards and self-discrepancies should arise in
transference along with concomitant differences in self-regulatory
focus, as predicted by self-discrepancy theory (Higgins, 1987).
Because self-discrepancy theory is, in part, an individualdifference model of the self and personality, evidence demonstrating an interface between our concept of the relational self and the
individual differences in self-standards and self-discrepancies it
specifies would further affirm the integrative nature of our theory
with respect to individual differences. Below we present data
supporting each of these predictions derived from our theory.
Changes in the content of the working self-concept. As we
have argued, when a significant-other representation is activated,
the working self-concept should be infused, in part, with knowledge reflecting the self-with-significant-other, or the relational self
(Hinkley & Andersen, 1996; see also Ogilvie & Ashmore, 1991).
In the research providing the most direct test of this hypothesis,
participants in the pretest session were asked to describe themselves, overall, as a baseline measure of their working selfconcept. Next, they were asked to describe a positive and a
negative significant other, and then the self they are with each of
these significant others—the latter serving as baseline measures of
the relational selves with each significant other.6 Several weeks
later in the experimental session, participants were presented with
descriptors about a new person who either did or did not resemble
their positive or negative significant other (depending on their
resemblance condition). They were then asked to generate descriptors to characterize themselves at that moment, as a measure of the
idiographic content of their working self-concept. They also classified each self-descriptor as positive or negative, as a measure of
idiographic self-evaluation.
Change in the working self-concept in transference was conceptualized in terms of a shift in listed content toward the relational
self. Hence, the degree of overlap between the descriptors of
participants’ working self-concept and those of the relational self
with that significant other was calculated, both at pretest and in the
experiment, so that shifts in degree of overlap could be examined.
As predicted, when the new person resembled participants’ own
significant other, thereby triggering transference, participants’
working self-concept shifted toward the relevant relational self, as
compared with participants in the no-resemblance condition (Hinkley & Andersen, 1996). This finding held for both positive and
negative significant others. The data demonstrate that significantother–self linkages are traversed when the significant-other representation is activated, rendering accessible those idiosyncratic
aspects of the self linked with this significant other. In the context
of transference, people become in part who they are with the
relevant significant other, verifying the claim that variability in
6
To effectively test the hypothesis that changes in self-definition and
self-evaluation occur in transference, we asked participants at pretest to
identify significant others with whom they typically behave and feel rather
differently than they do with others, and no participant reported having any
difficulty doing this.
THE RELATIONAL SELF
relational selves arises as a function of contexts that trigger
transference.
Self-evaluation in the working self-concept. As noted, content
shifts in the working self-concept in transference should be
accompanied by self-evaluative changes. Specifically, selfevaluation involving those aspects of the working self-concept
reflecting the relevant relational self should derive from the overall
affective tone of the significant-other representation, which embodies the relationship with the other as well. In the research just
described, this prediction was examined by summing the positive
and negative classifications that participants ascribed to each
working self-concept feature they listed during the experiment—
that also came to overlap with their relational-self descriptors. The
analyses of this sum showed that participants evaluated these
critical, overlapping working self-concept descriptors more positively (covarying out the same scores at pretest) when the new
person resembled their own positive, rather than negative, significant other. This effect was not seen in the no-resemblance conditions. Hence, self-evaluative shifts occur in the context of transference in the direction of the overall affect associated with the
significant other and the self– other relationship—positive shifts
when the significant other was positive and relatively negative
shifts when negative.
Self-regulation in the working self-concept. As predicted, the
pattern of self-evaluative changes described above was specific to
the transference condition and to the working self-concept descriptors that came to reflect (overlap with) the relevant relational self.
It was not observed for those aspects of the self that did not shift
toward the relational self (i.e., the nonoverlapping descriptors of
the working self-concept), and this may offer a window on selfregulatory processes in the working self-concept in response to
threat. Specifically, we argue that in the negative transference,
when participants encountered a new person who resembled a
negatively regarded significant other, the influx of negative
relational-self aspects into working memory is likely to have
constituted a threat to the self, which should thus elicit a selfregulatory response. In support of this, participants’ selfevaluations of their nonoverlapping working self-concept descriptors (again covarying out pretest evaluation) were by far the most
positive when the new person resembled one of their own negatively regarded significant others. Hence, self-evaluations of nonoverlapping self-concept descriptors showed a powerful reversal
of the hit taken by the self in this condition in the form of the
negative self-evaluative shifts shown in the overlapping selfdescriptors. That is, in the negative transference, an overwhelming
number of positively evaluated self-descriptors entered into the
working self-concept, directly contrasting away from the average
valence of the overlapping descriptors, bolstering the self in response. This can be understood in terms of compensatory selfenhancement—a self-regulatory response to self-threat in a negative transference encounter, arising from the influx of negative
self-aspects into the working self-concept.
Overall, this evidence demonstrates that the self in relation to
the significant other, defined idiographically, is activated in transference, such that the relational self becomes part of the working
self-concept. In short, it provides direct evidence for our overriding argument that the self is entangled with significant others such
that when transference occurs, relevant changes in the content of
the self occur accordingly. The data also show that transference
633
instigates self-evaluative changes in the working self-concept. For
self-aspects that overlap with the relational self, these changes
reflect the overall affect associated with the significant other and
the relationship. On the other hand, for nonoverlapping selfaspects in the case of a negative transference, a compensatory,
self-protective response emerges. Compensatory responses are
common in a variety of contexts (e.g., J. Greenberg & Pyszczynski, 1985; Steele, 1988; Taylor & Brown, 1988; Taylor & Lobel,
1989; Tesser, 1988), and presumably enable people to shore up
positive self-conceptions, restoring security and efficacious selfregulation when the resources or resiliency to do this are intact.
Furthermore, although one can argue that this evaluative contrasting away from the relational self might be a way of extricating
oneself from the entanglement with the negative significant other,
transference clearly does occur on the basis of negative significantother representations, and the content of the working self-concept
changes accordingly demonstrate the psychological importance of
negative transference.
Interpersonal roles, role violation, and affect. As we have
argued throughout, linkages between the self and the significant
other in memory are important in our theory because these linkages, when activated in transference, should elicit changes in how
one responds to others. We have also maintained that these linkages involve idiographic as well as normative knowledge reflecting typical patterns of relating with the other. To the degree that
such normative forms of knowledge are in fact activated when
idiographic significant-other representations are activated, it supports our assumption that these idiosyncratic representations from
one’s own individual life exist in a tangled web of associations
(see Cantor & Kihlstrom, 1987) with generic knowledge structures
of the sort more typically of interest to personality and social
psychologists.
Among the normative constructs that should be contained within
self-with-other knowledge are the interpersonal roles that the self
and the significant other occupy, which should help define the
self– other relationship (see Alexander & Higgins, 1993; Baldwin,
1992; Berscheid, 1994; Bugental, 1992, 2000; Clark & Reis, 1988;
Clark & Taraban, 1991; A. P. Fiske, 1992; Haslam, 1994; Mills &
Clark, 1994; Sarbin & Allen, 1968; Sedikides, Olsen, & Reis,
1993). Normative role relations—such as “expert”–“novice”
among parents and children, teachers and students, bosses and
employees—should also define expectations, goals, and feelings in
relationships, imbuing them with some of the meaning they ordinarily have (Carlson, 1981; Tomkins, 1979; see also Abelson,
1976, 1981; Schank & Abelson, 1977). Building on this idea, we
have examined the affective consequences of experiencing a violation of expectations based on the role relationship with a significant other in the context of transference (Andersen & Baum,
1999).
Specifically, if the role relationship one shares with a positively
regarded significant other is activated when the representation of
the other is activated with a new person, a role violation exhibited
by this new person should violate role-based expectations and
thereby disrupt the positive affect that would otherwise ensue. The
fulfillment or frustration of motives within a particular role
relationship clearly influences affective experience (e.g.,
Carver & Scheier, 1998; Martin, Tesser, & McIntosh, 1993; see
also Oatley & Bolton, 1985). Hence, when a new person who
resembles a positive significant other occupies a role that is
634
ANDERSEN AND CHEN
incongruent with the role typically occupied by the significant
other, this should produce negative mood because it signals that
positive outcomes usually pursued will not be attained. By
contrast, when there is role congruence in transference, affective experiences should reflect the overall affect associated with
the significant other.
In the pretest session of research testing these predictions,
participants described a positively regarded significant other who
was an authority figure to them. Then, in the experimental session,
they were exposed to descriptors about a new person who either
did or did not resemble this significant other. They were also told
that this new person was either an expert in a task to be done
together, thereby matching the authority role of their significant
other, or a novice in the task, thereby mismatching their significant
other’s role. After reading the descriptors about the new person,
participants completed measures of self-reported mood. As predicted, the results showed that, in the context of transference, role
incongruence—vis-à-vis the role of the significant other—led participants to report a negative, depressed mood state as compared
with the relatively more positive mood reported by participants
experiencing role congruence. This effect did not occur in the
absence of significant-other resemblance, indicating that the effect
was specific to the transference condition—when the role relationship should have been activated on the basis of activation of the
significant-other representation.
This evidence suggests that normative knowledge reflecting
the relational self—in this case, normatively defined self– other
roles—is activated on the basis of significant-other resemblance
in a new person. In transference, then, participants appear to
have been prepared to play the same role they play (or played)
with their positive significant other, so that when the role
relationship was violated, negative affect arose. These data
showing negative affect in a positive transference also indicate
that affect in transference does not always parallel evaluation
and that not all negative affect in transference elicits a selfprotective, compensatory response.7
Individual differences in self-discrepancy and affective vulnerability. Another normative element of self– other relationships
that should be represented in memory as part of linkages between
self and significant others is the set of standards that significant
others hold for oneself. According to self-discrepancy theory (Higgins, 1987), people’s beliefs about what their significant others
hope for them (ideals) or feel it is their duty or obligation to be
(oughts) may be discrepant from how these others actually view
them, that is, view the actual self. And when ideal standards are
discrepant from the actual self, there is a vulnerability to depression and dejection-related affect, whereas when ought standards
are discrepant from the actual self, there is a vulnerability to
agitation-related affect.
The interface between self-discrepancy theory and our model
was examined by preselecting participants on the basis of whether
they had an ideal or an ought self-discrepancy (and not both) from
the standpoint of one of their parents (using the Selves Questionnaire; Higgins et al., 1986). In all conditions, participants indicated
they liked or loved the parent. In the experiment, ideal- and
ought-discrepant participants were randomly assigned to a condition in which they learned about a new person who did or did not
resemble their own parent (with no explicit reference to
discrepancy-relevant information). This enabled a straightforward
test of our hypothesis that the relevant self-discrepancy should
be activated when the significant-other representation is activated— on the basis of self-reported mood.
As predicted, ideal-discrepant participants showed more depressed mood when the new person resembled their own parent
than they did in the no-resemblance condition. This effect did not
occur among ought-discrepant participants. These data demonstrate that ideal self-discrepancies are indirectly activated among
ideal-discrepant participants in the context of transference—that is,
in the context that activated the mental representation of the parent
in relation to whom they have the discrepancy, resulting in the
predicted affective consequences.8
Evidence for the activation of ought discrepancies in transference was also found. Specifically, ought-discrepant participants
showed more resentful and hostile mood when the new person
resembled their own parent than they did in the no-resemblance
condition (on resentment, see also Strauman & Higgins, 1988).
This effect did not occur among ideal-discrepant participants,
which suggests the indirect activation of ought self-discrepancies
in transference. Beyond this, ought-discrepant participants in the
resemblance condition also reported feeling less calm than they did
in the control condition—as assessed right after they were informed that they would not actually be meeting with the new
person, when all other participants were relieved and relatively
calm. Again, no such effect occurred for ideal-discrepant participants.9
7
Although it is unclear exactly why compensatory responses did not
emerge in this research on role violation, it may be that participants had
little practice in advance in dealing with role reversals with their significant
others (and the negative affective consequences of such a role violation)
and thus little practice with coping with them in a compensatory manner.
This stands in contrast to the more commonly experienced forms of threat
to self or relationship reported earlier. Prior practice with the threat may
thus account for compensatory effects in transference, a matter that warrants future research attention.
8
Although this effect did not occur in an additional condition when an
explicit priming procedure was used, the effect reported still holds. The
explicit prime involved statements about what the new person looks for in
friends—in the ideal condition, friends that live up to his or her ideals, and
in the ought condition, to his or her sense of duties and obligations.
Although we do not know exactly what transpired in the priming condition,
the fact that it so thoroughly dampened the reported effects raises interesting questions about the degree to which a new relationship that
brings with it a new ideal standard might on occasion inhibit an ideal
discrepancy associated with a past significant other, even when this
significant-other representation is in fact activated in the new relationship. This is a speculative but feasible outcome. This notion also
comports well with some evidence suggesting that new standards arising when one becomes a parent, for example, can override past standards (Alexander & Higgins, 1993). Thus, there is much more to be
examined beyond the simple process of a new person activating a prior
significant-other representation associated with a past discrepancy.
9
This effect was captured because the mood measures were administered twice, as is typical in our studies, so that we can compare perceptions
when the interaction is anticipated with those when it is not. Although we
rarely find differences as a function of administration time, and effects
observed have tended to concern self-reported evaluation (Andersen &
Baum, 1994; Andersen et al., 1996; Baum & Andersen, 1999), both
administration times are typically included in order to detect any variation
in responding based on anticipating (or not) an imminent interaction.
THE RELATIONAL SELF
Individual differences in self-discrepancy and self-regulatory
focus. In self-discrepancy theory, the affective consequences that
arise when ideal and ought self-discrepancies are activated are
seen to be part of a broader self-regulatory system (Higgins,
1996a, 1996b). That is, ideal standards elicit a self-regulatory
focus on positive outcomes, whereas ought standards elicit a focus
on negative outcomes. It is precisely because ideal standards lead
to a focus on the promotion of positive outcomes that activating
discrepancies from these standards, which signals a failure to attain
positive outcomes, produces dejection-related affect. Similarly, it
is because ought standards orient the self toward preventing negative outcomes that agitation-related affect ensues when discrepancies from these standards are activated, which signals a failure to
avoid negative outcomes. If the consequences of activating
self-discrepancies ultimately reflect particular self-regulatory
systems, the self-regulatory implications of activating selfdiscrepancies from a significant other’s standpoint should emerge
in transference.
To test this hypothesis, participants in the same research were
asked to complete a measure of motivation. The predictions were
that in the transference condition, ideal-discrepant participants, for
whom a promotion self-regulatory focus should be activated,
should exhibit greater approach tendencies—in the form of being
more open and proactive about engaging with the new person—
whereas ought-discrepant participants, for whom a prevention
self-regulatory focus should be salient, should show greater avoidance tendencies. The data were in line with these predictions.
Specifically, in the context of transference, ideal-discrepant individuals experienced considerably less motivation to avoid the
person when they were anticipating that they would meet this
person as compared with after they were informed they would not
actually meet him or her (at which point promotion was no longer
relevant), whereas ought-discrepant individuals experienced more
avoidance motivation while anticipating the meeting than after
learning it would not occur (at which point prevention was no
longer relevant). This asymmetrical approach–avoidance pattern
emerged only in the transference condition, suggesting that distinct
self-regulatory systems were likely activated among ideal- and
ought-discrepant individuals, respectively, on the activation of the
relevant significant-other representation.
By capturing the role of individual differences in selfdiscrepancies in the phenomenon of transference, these data forge
a substantive link between our theory and a more traditional
individual-differences approach. Although the methods and measures of self-discrepancy theory are not the most traditional, ideal
and ought standards are normatively defined in terms of categorical individual differences, even though their exact content is
assumed to be idiographic. Integrating normative self-standards
and self-discrepancies into our model therefore enables us to tread
on more traditional ground in our effort to demonstrate the relevance of the relational self to personality.
Summary. To summarize, this second body of evidence shows
that activation of a significant-other representation instigates shifts
in the self toward the relevant relational self. These shifts involve
idiosyncratic changes in the content of the working self-concept, as
well as changes in self-evaluation that parallel the positivity or
negativity of the self– other relationship—that is, positive selfevaluation arises in positive transference and negative selfevaluation in negative transference. The data also show that the
threat to the self that occurs in a negative transference elicits
635
compensatory self-enhancement, with the overall working selfconcept taking on a dramatically more positive self-evaluation
under these circumstances than in any other condition. Hence,
self-protective self-regulation occurs. Although one might argue
that this self-evaluative contrast away from the self-with-thesignificant-other reflects extricating from the entanglement with
this negative significant other, the evidence also shows negative
changes in self-evaluation and clearly demonstrates that negative
transference occurs. In our view, the fact that this overall positive
shift in self-evaluation in the negative transference was the most
pronounced observed in the study, even though counter to the
valence of the transference, indicates a profound reactivity within
the entanglement rather than, in effect, disentanglement.
The evidence also demonstrates that activating a significantother representation in transference activates normative relational
knowledge associated with the significant other. Role information,
in particular, is activated and has implications for how interactions
are experienced. That is, with a new person in transference, a role
violation results in disappointment and negative affect, presumably
because it signals that goals and expectations normally linked to
the role will not be realized. These findings show that negative
affect can arise in a positive transference and that self-regulation in
response does not always occur (or does not occur successfully) on
the basis of such negative experiences.
Normative standards or prescriptive assumptions that the significant other tends to hold are also activated in transference. Hence,
self-discrepancies held from a parent’s standpoint are elicited
when the mental representation of this parent is activated. As a
result, nuanced affective responses emerge, as do shifts in selfregulatory focus, as predicted by self-discrepancy theory. In transference, dejection-related affect occurs among people with ideal
self-discrepancies, and agitation-related affect occurs among those
with ought self-discrepancies. Moreover, shifts in the motivation
to be emotionally close to the new person occur that reflect the
relevant self-regulatory focus. In short, activation of the parental
representation results not only in activation of the self one is with
the significant other but also in activation of knowledge about
what the significant other wants from one and about how he or she
views one’s actual self.
Implications of the Evidence for Relational Selves as an
If–Then Model of Personality
Taken together, the two bodies of findings we have presented
provide evidence for our key propositions regarding the relational
self. They support our view that, by virtue of the emotional and
motivational significance of significant others to the self, linkages
are formed in memory between self and significant-other representations. Each linkage establishes a relational self, which captures the version of the self one is in relation to that significant
other. When the significant-other representation is activated, this
activation spreads to the associated relational self, so that the
working self-concept becomes infused with aspects of this particular self, and one’s evaluation of these self-aspects in turn reflects
the overall evaluation of the self– other relationship along with
self-regulatory shifts in self-evaluation. In addition to idiographic
changes in self-definition and self-evaluation, the effects of transference also involve normative self-aspects such as interpersonal
roles and self-standards that lead to shifts in affect, motivation, and
636
ANDERSEN AND CHEN
self-regulation—all reflecting the self with the significant other,
even though the other is not there.
In our view, the data clearly demonstrate the emergence of
relational selves in the context of transference and imply that
variability in the self is determined in part by the contextual
activation and use of significant-other representations. It is in this
basic sense that we argue that our theory constitutes an if–then
model of personality, paralleling the broader social– cognitive,
interactionist model developed by Mischel and Shoda (1995). As
described earlier, Mischel and Shoda argue that the if in each
if–then relation activates an idiographic set of cognitive–affective
units (e.g., encodings, feelings, goals), which gives rise to the
psychological situation that then elicits the then in that relation.
If–then relations are seen as the basic units of personality, and an
individual’s overall set of if–then relations is seen as his or her
unique personality signature.
In our more targeted social– cognitive framework, ifs are interpersonal situations in which a new person, by virtue of his or her
resemblance to a significant other, activates the representation of
this significant other, and thens are the constellation of responses
reflecting the associated relational self that then emerges in relations with the new person. Thus, relational selves are the
cognitive–affective units, which contain both idiographic and normative elements, that give rise to the psychological situations that
underlie the broad range of thens we have measured in our research. In terms of an individual’s overall signature, we would
argue that if–then relations involving relational selves are a major
source of the interpersonal elements of the individual’s personality. Like the broader if–then model, our theory thus allows for
variability and stability in personality responding. Variability lies
in the activation of different relational selves across interpersonal
contexts in which transference is triggered, whereas stability derives from the steady influence of chronically accessible
significant-other representations and associated relational-self
knowledge as well as from the predictability of the emergence
of relational selves in the social– cognitive phenomenon of
transference.
Beyond its fit with social– cognitive, interactionist views of
personality, in particular Mischel and Shoda’s (1995) model, we
have argued that our theory of relational selves constitutes a model
of personality by virtue of its motivational underpinnings, which
are basic to most broad-based theories of personality (e.g., Sullivan, 1953). In addition, our theory encompasses both idiographic
and nomothetic elements, thereby forging simultaneous links to
idiographic approaches to personality and to approaches that treat
personality in terms of nomothetic individual differences. Finally,
our theory incorporates self-regulatory processes, which are
widely considered to be basic to personality functioning (e.g.,
Mischel et al., 1996).
Related Literatures: Parallels and Implications
Our theory of the relational self and personality dovetails nicely
with recent developments in several research literatures. We
briefly discuss some of the implications of these developments for
our theory.
Multiple levels of the self. Our theory is compatible with a
variety of conceptual frameworks that share the critical assumption
that the self is defined in part in relation to others (e.g., Aron et al.,
1991; Baumeister & Leary, 1995; Breakwell, 1992; Brewer, 1991;
Brewer & Gardner, 1996; Crocker, Luhtanen, Blaine, & Broadnax,
1994; Deaux, 1992, 1993; Gaertner, Sedikides, & Graetz, 1999;
Higgins, 1987, 1989a; Markus & Cross, 1990; Moretti & Higgins,
1999; E. R. Smith et al., 1999; E.R. Smith & Henry, 1996; Tajfel
& Turner, 1979; Tesser, 1988; Triandis, 1989; Turner, Oakes,
Haslam, & McGarty, 1994; Wright, Aron, McLaughlin-Volpe, &
Ropp, 1997; see also Andersen et al., 1997). Despite the common
focus on the social nature of the self, these frameworks vary in
terms of the level or levels at which the self is conceptualized. For
example, research in the tradition of social identity theory has
focused on distinguishing personal and social identities, with the
former referring to the self as an autonomous entity and the latter
referring to the selves experienced as group memberships (e.g.,
Brewer, 1991; Crocker & Luhtanen, 1990; Deaux, 1992, 1993;
Trafimow, Triandis, & Goto, 1991; Triandis, 1989; Turner et al.,
1994). A newer set of theories has focused on the self in relation
to specific individuals, such as significant others (e.g., Aron et al.,
1991; Baldwin, 1992; S. Chen & Andersen, 1999).
Our concept of the relational self is obviously most closely akin
to the latter literature. However, elsewhere we have suggested that
theories focused on different levels of self-definition can be integrated in ways that include personal identity as well as the self in
relation both to other individuals and to groups (see Andersen et
al., 1997). From such a perspective, the self in relation to significant others would offer one level of analysis intermediate in
inclusiveness between personal and social identities. Recent theoretical frameworks have, in fact, conceptualized identity in exactly
this way (e.g., Brewer & Gardner, 1996; Sedikides & Gaertner,
2001).
Considering the connection between our concept of the relational self and other levels of self-definition raises a host of
provocative issues. For example, it raises questions about the kinds
of internal and external factors that determine which level of
self-definition is operative at a given moment (e.g., Brewer &
Gardner, 1996; Brewer & Weber, 1994; Simon, Pantaleo, & Mummendey, 1995; Trafimow, Silverman, Fan, & Law, 1997), and
about the cues that are most likely to activate one level over
another (e.g., Simon et al., 1995; Trafimow et al., 1991). Also
interesting to consider are the ways in which different levels of
self-definition may interact with one another (e.g., Y. R. Chen,
Brockner & Katz, 1998; Seta & Seta, 1996; Tesser, 1988) and
whether different levels reflect separate cognitive systems (e.g.,
Trafimow et al., 1991) or an integrated system from which different selves draw (e.g., Reid & Deaux, 1996). All of these questions
are relevant to research on relational selves as well as to research
on social identities. For example, if social identities are defined in
part by specific relationships, multiple levels of identity may well
be active simultaneously, and this warrants examination.
The emerging integration of multiple levels of self-definition
into one framework raises interesting questions and issues regarding self-regulation as well. For example, future research needs to
address likely differences between self-regulation in the service of
the personal self—that is, of “me” and “mine”—from selfregulation in the service of social selves—that is, of “we” and
“ours” (see Higgins, 1996d). Although personal self-regulation, as
we have shown, clearly can be profoundly social in that significant
others are of pressing importance in determining the apparent
necessity for self-protective self-regulation, this is not self-
THE RELATIONAL SELF
regulation in the service of group or collective goals but instead in
the service of personal goals. Varying the referent—the personal,
relational, or collective self—at issue in self-regulation may thus
call into play both different means and different ends for selfregulation, and this is worthy of serious examination, because our
work shows that relationship-protective self-regulation arises
when the referent is the relationship. In addition, even though we
have examined personal self-regulation as it relates to possible and
actual selves from a significant other’s standpoint, we have yet to
distinguish personal self-regulation of this kind, that may be encumbered by the significant other, from that which is more intrinsic or self-determined, even if relational (e.g., Deci, 1995). Overall, future research examining such issues has the potential to
further the conceptual integration of existing models of the self.
The relational self, culture, and gender. In some models of the
self, possible cross-cultural, subcultural, and gender-based variations in self-definition have been emphasized (e.g., Cross & Madson, 1997; Markus & Cross, 1990; Markus & Kitayama, 1991;
Rhee, Uleman, Lee, & Roman, 1995; Triandis, 1989). Overall,
these have argued that the degree to which the self is interdependent and bound up with others, as compared with the degree to
which the self is independent and separate, varies as a function of
culture and gender. Although we concur that interdependence and
independence may vary across cultures and subcultures, and perhaps also gender, we still assume that the self is fundamentally
entangled with significant others, to some extent, for all people.
Regardless of culture or gender, the self and personality should be
defined in part by a person’s idiosyncratic set of relational selves.
We also suggest that, rather than characterize such variations in
terms of overall interdependence or independence (e.g., Markus &
Kitayama, 1991; Triandis, 1988), it may be appropriate to consider
such variations more specifically in terms of levels of selfdefinition—namely, whether the self is defined in relation to
another individual or in relation to a social group, both of which
would reflect a kind of interdependence. Because the self very
likely exists at multiple levels, cross-cultural, subcultural, and
gender differences in the self are likely to depend on the level of
analysis at which they are conceived. Differences across Western
and Eastern cultures in the extent to which the self is bound up
with group memberships may not always parallel the degree to
which the self is bound up with interpersonal relationships. Moreover, the individuality of personal identity emphasized in Western
culture clearly does not preclude the self from being bound up with
relationships to some degree, as we have shown in our work, nor
of being linked to group memberships and social identities, as
other research has suggested. Similarly, although there may be
clear cross-cultural differences in the baseline prevalence of different ways of characterizing the self, there are also clear crosscultural and subcultural variations in the prevalence of various
kinds of contexts that render some aspects of the self operative
relative to others. Cultural differences in the prevalence of contexts
that evoke particular selves thus coexist with cultural differences
in baseline perceptual tendencies (Kitayama, Markus, Matsumoto,
& Norasakkunkit, 1997). Thus, differing contextual triggers can be
teased apart from differing perceptions, and this offers a significant
advance in research on cultural influences on self and identity, one
also present in our paradigm. Hence, examining relational selves
cross-culturally seems an increasingly promising line of inquiry.
637
It is our view that for most people, the self is in fact entangled
at both the interpersonal level with specific others in one’s life and
at the collective level in relation to the social groups with which
one identifies. At a cognitive level, this implies that linkages exist
in memory between significant-other representations and the self,
as well as between knowledge designating social categories and
the self. These linkages essentially represent different forms of
self-with-other knowledge, as well as differences in how idiosyncratic or normative the knowledge is. Considering the cognitive
underpinnings of the social self in this way suggests that cultureand gender-based variations in the social self may reflect differences in the accessibility of different forms of self-with-other
knowledge or in the prevalence of triggering contexts, or both,
rather than differences in the availability of different forms per se
(Brewer & Gardner, 1996; Brewer & Roccas, 2001; Higgins,
1989b; Higgins & May, 2001; Kitayama et al., 1997; Reid &
Deaux, 1996; Trafimow et al., 1991).
Overall, then, we speculate that few, if any, limiting conditions
are likely for the existence of the relational self. Even in highly
individualist settings, such as those of American society, the self is
not entirely independent. In this light, we find intriguing those data
that suggest that the individual self has “motivational primacy”
over the collective self (for a review, see Sedikides & Gaertner,
2001; see also Gaertner et al., 1999; Simon et al., 1995; cf. Y. R.
Chen et al., 1998), in that threats to the individual self elicit more
severe, negative reactions, for which the collective self can serve
as a self-protective resource, than do similar threats to the collective self, which do not lead people to take recourse in bolstering or
protecting the individual self (Gaertner et al., 1999). Although we
take no stand on the primacy issue vis-à-vis the relational self, we
assume the fundamental point to be that the individual self and
relational selves coexist, even if one were to dominate. It is worth
noting that, to our knowledge, work on the primacy of the individual self has yet to examine relational selves. We speculate that
given the ubiquity of the transference phenomenon and the breadth
of its consequences for the self and personality, the self in relation
to significant others may carry considerable motivational weight,
perhaps approaching, if not matching, the individual self. Nonetheless, we hold to the view that a model that integrates the
multiple levels at which the self may exist in relation to others will
ultimately be most fruitful (see Brewer & Roccas, 2001).
Finally, we underscore that we do not wish to imply in our
model that any personal experiences not wrapped up with one’s
significant others are irrelevant to the self or to personality functioning. Indeed, considerable evidence indicates that private, internal states are important in self-perception and self-judgment
(e.g., Andersen, 1984, 1987; Andersen, Lambert, & Dick, 1999;
Andersen & Williams, 1985; Levine, Wyer, & Schwartz, 1994;
Schwarz, 1990; see also Deci & Ryan, 1985). This evidence
suggests that people take their own covert, subjective experiences
very seriously indeed, by defining themselves in terms of internal,
covert interpretations, feelings, and wishes. The subjective nature
of construal and responding is a given in social psychology.
However, in our view, this need not imply either the centrality of
the individual, personal self, or a lack of importance of relations
with others. In fact, significant others are essential in the formation
of people’s private, internal responses and the various constructs
they ultimately use to interpret their experience (see also Kelly,
1955), implying an interplay between individual and interpersonal
ANDERSEN AND CHEN
638
knowledge (e.g., Hardin & Higgins, 1996; Wegner, 1995) in the
development of self (Harter, 1999; Harter, Waters, & Whitesell,
1998).
Concluding Remarks
We have presented an interpersonal social– cognitive theory of
the self and personality that is grounded in the phenomenon of
transference. The theory posits that individuals possess multiple
selves in relation to the various significant others in their lives,
each linked in memory with a particular significant other. It is
because linkages exist between the self and significant-other representations that when a significant-other representation is activated in an encounter with a new person, associated self-withsignificant-other knowledge is set into motion. As a result, the
relevant relational self comes into play, as manifested in the
individual’s emotional, motivational, and behavioral responses in
the interpersonal encounter. Different relational selves thus unfold
dynamically across interpersonal contexts. It is in this regard that
our view of relational selves can be conceptualized in if–then
terms, whereby ifs are interpersonal situations in which transference is triggered, and thens are the manifestations of the relational
self that emerge in these situations. We contend that an individual’s overall repertoire of relational selves is an influential source of
his or her interpersonal patterns and, as such, of his or her personality. Our theory of the relational self can thus be seen as an
everyday instantiation of the broader, social– cognitive viewpoint
on personality.
Overall, we propose a conceptualization of the self and personality that is fundamentally interpersonal—focusing on the self in
relation to specific significant others. Although we do not mean to
imply that there is nothing to the self beyond the interpersonal,
or that there are no cultural or other differences in the extent to
which significant-other relationships shape the self, we regard
significant-other relationships as basic to self-experience and argue that they provide the contexts in which much of selfknowledge is derived and thus serve as guideposts for selfdefinition and self-evaluation in memory. Of course, many
pressing questions are yet to be addressed in this research involving, for example, any cultural or subcultural differences that may
exist, possible troublesome or pathological consequences, and
even the most positive of consequences that may emerge, such as,
perhaps, a greater capacity for empathy when a new person reminds one of a significant other about whom one cares or for
whom one has cared. But the data clearly suggest that basic needs
relevant to one’s significant other arise in transference, as does
self-regulation.
The evidence also clearly shows that transference occurs for
both positive and negative significant others. Moreover, because
self-protective self-regulation is provoked by a negative transference, presumably to ward off the threat to the self it poses, this
suggests the evocative emotional power of the negative transference context. Our pursuit of questions about differences between
negative and positive transference experiences, of course, may
lead one to wonder if there is not something rather artificial about
classifying significant-other representations as positive or negative. All significant-other representations are likely to be fairly
complex and may contain numerous contradictory aspects, just as
one’s self-representations do (Sande, Goethals, & Radloff, 1988).
Hence, we can assume this complexity may translate into some
amount of ambivalence for many significant others, suggesting
that subtle variations in where one places a significant other on the
valence continuum may well matter somewhat less than whether or
not the person is (or has been) significant. In fact, the emotional
suffering one may experience in transference can clearly arise in a
positive transference, just as it can in a negative transference—for
example, when a positive significant other is linked to selfdiscrepancies or when role violations are experienced with a new
person. In short, we believe that this model and the evidence have
implications both for vulnerability and for resilience in the relational self (Andersen et al., 2002).
On another level, future work would do well to examine our
underlying assumption that an individual’s repertoire of
significant-other representations and self-with-significant-other
knowledge can be expanded as new people become significant.
The evidence suggests that people can form new significant-other
relationships with people beyond their family of origin and that
transference occurs on the basis of a wide variety of significantother representations. Presumably when one feels fondness toward
someone and begins to invest emotionally and motivationally in
the person, in small and then in larger ways, one sets the stage for
forming a significant-other representation designating this person.
New aspects of self are likely to then be developed or enhanced on
the basis of the new relationship, perhaps even self-aspects not
previously operative with any prior significant other. This possibility implies that the self can be extended in positive directions on
the basis of newly formed relationships and thus offers some hope
for changing counterproductive patterns and building desired identities—with new friends, colleagues, mentors, or romantic
partners.
In conclusion, the relational self and personality are embodied,
at least in part, in transference—a phenomenon that tracks idiographic variability in the self and personality, while also reflecting
continuity, in accordance with well-established social– cognitive
principles. Transference occurs by means of the activation of
mental representations of significant others, which accordingly
evokes the associated self-with-significant-other— or the relational
self. This system of cognitive–affective units, composed of
significant-other and relational-self knowledge, defines variability
and stability in our if–then model of the relational self and personality. It suggests that significant others are crucial to selfdefinition and its vicissitudes as well as to affective and motivational experience, self-regulation, and interpersonal behavior.
References
Abelson, R. P. (1976). Script processing in attitude formation and decision
making. In J. S. Carroll & J. W. Payne (Eds.), Cognition and social
behavior (pp. 33– 45). Hillsdale, NJ: Erlbaum.
Abelson, R. P. (1981). The psychological status of the script concept.
American Psychologist, 36, 715–729.
Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. D. (1978). Learned
helplessness in humans: Critique and reformation. Journal of Abnormal
Psychology, 87, 49 –74.
Adler, A. (1957). Understanding human nature. New York: Fawcett Premier. (Original work published 1927)
Ainsworth, M. D. S., Blehar, M. C., Walters, E., & Wall, S. (1978).
Patterns of attachment: A psychological study of the strange situation.
Hillsdale, NJ: Erlbaum.
THE RELATIONAL SELF
Alexander, M. J., & Higgins, E. T. (1993). Emotional trade-offs of becoming a parent: How social roles influence self-discrepancy effects. Journal
of Personality and Social Psychology, 65, 1259 –1269.
Allport, G. (1937). Personality: A psychology interpretation. New York:
Holt, Rinehart & Winston.
Andersen, S. M. (1984). Self-knowledge and social inference: II. The
diagnosticity of cognitive/affective and behavioral data. Journal of Personality and Social Psychology, 46, 294 –307.
Andersen, S. M. (1987). The role of cultural assumptions in self-concept
development. In K. Yardley & T. Honess (Eds.), Self and identity:
Psychological perspectives (pp. 231–246). New York: Wiley.
Andersen, S. M., & Baum, A. B. (1994). Transference in interpersonal
relations: Inferences and affect based on significant-other representations. Journal of Personality, 62, 4, 460 – 497.
Andersen, S. M., & Bem, S. L. (1981). Sex typing and androgyny in dyadic
interaction: Individual differences in responsiveness to physical attractiveness. Journal of Personality and Social Psychology, 41, 74 – 86.
Andersen, S. M., & Berk, M. S. (1998). Transference in everyday experience: Implications of experimental research for relevant clinical phenomena. Review of General Psychology, 2, 81–120.
Andersen, S. M., Chen, S., & Miranda, R. (2002). Significant others and
the self. Self and Identity, 1, 159 –168.
Andersen, S. M., & Cole, S. W. (1990). “Do I know you?”: The role of
significant others in general social perception. Journal of Personality
and Social Psychology, 59, 384 –399.
Andersen, S. M., & Glassman, N. S. (1996). Responding to significant
others when they are not there: Effects on interpersonal inference,
motivation, and affect. In R. M. Sorrentino & E. T. Higgins (Eds.),
Handbook of motivation and cognition (Vol. 3, pp. 262–321). New
York: Guilford Press.
Andersen, S. M., Glassman, N. S., Chen, S., & Cole, S. W. (1995).
Transference in social perception: The role of chronic accessibility in
significant-other representations. Journal of Personality and Social Psychology, 69, 41–57.
Andersen, S. M., Glassman, N. S., & Gold, D. (1998). Mental representations of the self, significant others, and nonsignificant others: Structure
and processing of private and public aspects. Journal of Personality and
Social Psychology, 75, 845– 861.
Andersen, S. M., & Klatzky, R. L. (1987). Traits and social stereotypes:
Levels of categorization in person perception. Journal of Personality
and Social Psychology, 53, 235–246.
Andersen, S. M., Lambert, L., & Dick, W. (1999). Significant-other examplars: Processing efficiency in instance-based judgments. Manuscript
submitted for publication.
Andersen, S. M., Reznik, I., & Chen, S. (1997). Self in relation to others:
Cognitive and motivational underpinnings. In J. G. Snodgrass & R. L.
Thompson (Eds.), The self across psychology: Self-recognition, selfawareness, and the self-concept (pp. 233–275). New York: New York
Academy of Science.
Andersen, S. M., Reznik, I., & Manzella, L. M. (1996). Eliciting transient
affect, motivation, and expectancies in transference: Significant-other
representations and the self in social relations. Journal of Personality
and Social Psychology, 71, 1108 –1129.
Andersen, S. M., & Williams, M. (1985). Cognitive/affective reactions in
the improvement of self-esteem: When thoughts and feelings make a
difference. Journal of Personality and Social Psychology, 48, 1086 –
1097.
Aron, A., & Aron, E. N. (1996). Self and self-expansion in relationships.
In G. J. O. Fletcher & J. Fitness (Eds.), Knowledge structures in close
relationships: A social psychology approach (pp. 325–344). Mahwah,
NJ: Erlbaum.
Aron, A., Aron, E. N., & Smollen, D. (1992). Inclusion of other in the self
scale and structure of interpersonal closeness. Journal of Personality and
Social Psychology, 63, 596 – 612.
639
Aron, A., Aron, E. N., Tudor, M., & Nelson, G. (1991). Close relationships
as including other in the self. Journal of Personality and Social Psychology, 60, 241–253.
Aron, A., & Fraley, B. (1999). Relationship closeness as including other in
the self: Cognitive underpinnings and measures. Social Cognition, 17,
140 –160.
Ashmore, R. D, & Ogilvie, D. M. (1992). He’s such a nice boy . . . when
he’s with Grandma: Gender and evaluation in self-with-other representations. In T. M. Brinthaupt & R. P. Lipka (Eds.), The self: Definitional
and methodological issues (pp. 236 –290). Albany: State University of
New York Press.
Ayduk, O., Mendoza-Denton, R., Mischel, W., Downey, G., Peake, P. K.,
& Rodriguez, M. (2000). Regulating the interpersonal self: Strategic
self-regulation for coping with rejection sensitivity. Journal of Personality and Social Psychology, 79, 776 –792.
Bacon, M. K., & Ashmore, R. D. (1985). How mothers and fathers
categorize descriptions of social behavior attributed to daughters and
sons. Social Cognition, 3, 193–217.
Bakan, D. (1966). The duality of human existence. Chicago: Rand McNally.
Baldwin, M. W. (1992). Relational schemas and the processing of information. Psychological Bulletin, 112, 461– 484.
Baldwin, M. W., Carrell, S. E., & Lopez, D. F. (1990). Priming relationship
schemas: My advisor and the Pope are watching me from the back of my
mind. Journal of Experimental Social Psychology, 26, 435– 454.
Baldwin, M. W., Fehr, B., Keedian, E., Seidel, M., & Thomson, D. W.
(1993). An exploration of the relational schemata underlying attachment
styles: Self-report and lexical decision approaches. Personality and
Social Psychology Bulletin, 19, 746 –754.
Baldwin, M. W., Keelan, J. P. R., Fehr, B., Enns, V., & Koh-Rangarajoo,
E. (1996). Social– cognitive conceptualization of attachment working
models: Availability and accessibility effects. Journal of Personality
and Social Psychology, 71, 94 –109.
Baldwin, M. W., & Sinclair, L. (1996). Self-esteem and “if . . . then”
contingencies of interpersonal acceptance. Journal of Personality and
Social Psychology, 71, 1130 –1141.
Banaji, M. R., & Prentice, D. A. (1994). The self in social context. Annual
Review of Psychology, 45, 297–332.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral
change. Psychological Review, 84, 191–215.
Bandura, A. (1986). Social foundations of thought and action: A social
cognitive theory. Englewood Cliffs, NJ: Prentice Hall.
Bandura, A. (1989). Human agency in social– cognitive theory. American
Psychologist, 44, 1175–1184.
Bargh, J. A. (1990). Auto-motives: Preconscious determinants of social
interaction. In E. T. Higgins & R. M. Sorrentino (Eds.), Handbook of
motivation and cognition: Foundations of social behavior (Vol. 2, pp.
93–130). New York: Guilford Press.
Bargh, J. A. (1997). The automaticity of everyday life. In R. S. Wyer (Ed.),
The automaticity of everyday life: Advances in social cognition (pp.
1– 61). Mahwah, NJ: Erlbaum.
Bargh, J. A. (1999). The cognitive monster: The case against the controllability of automatic stereotype effects. In S. Chaiken & Y. Trope (Eds.),
Dual-process theories in social psychology (pp. 361–382). New York:
Guilford Press.
Bargh, J. A., Bond, R. N., Lombardi, W. J., & Tota, M. E. (1986). The
additive nature of chronic and temporary sources of construct accessibility. Journal of Personality and Social Psychology, 50, 869 – 878.
Bargh, J. A., & Gollwitzer, P. M. (1994). Environmental control of goaldirected action: Automatic and strategic contingencies between situations and behavior. In W. D. Spaulding (Ed.), Integrative views of
motivation, cognition, and emotion (pp. 71–124). Lincoln: University of
Nebraska Press.
Bargh, J. A., & Thein, R. D. (1985). Individual construct accessibility,
640
ANDERSEN AND CHEN
person memory, and the recall–judgment link: The case of information
overload. Journal of Personality and Social Psychology, 49, 1129 –1146.
Bartholomew, K., & Shaver, P. R. (1998). Methods of assessing adult
attachment: Do they converge? In J. A. Simpson & W. S. Rholes (Eds.),
Attachment theory and close relationships (pp. 25– 45). New York:
Guilford Press.
Batson, C. D. (1990). How social an animal? The human capacity for
caring. American Psychologist, 45, 336 –346.
Baum, A., & Andersen, S. M. (1999). Interpersonal roles in transference:
Transient mood states under the condition of significant-other activation.
Social Cognition, 17, 161–185.
Baumeister, R. F. (1991). Meaning of life. New York: Guilford Press.
Baumeister, R. F. (1997). Esteem threat, self-regulatory breakdown, and
emotional distress as factors in self-defeating behavior. Review of General Psychology, 1, 145–174.
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for
interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117, 497–529.
Becker, E. (1971). The birth and death of meaning (2nd ed.). New York:
Free Press.
Becker, E. (1973). The denial of death. New York: Free Press.
Berk, M. S., & Andersen, S. M. (2000). The impact of past relationships on
interpersonal behavior: Behavioral confirmation in the social-cognitive
process of transference. Journal of Personality and Social Psychology, 79, 546 –562.
Berscheid, E. (1994). Interpersonal relationships. Annual Review of Psychology, 45, 79 –129.
Bombar, M., & Littig, L., Jr. (1996). Babytalk as a communication of
intimate attachment: An initial study in adult romances and friendships.
Personal Relationships, 3, 137–158.
Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. New York:
Basic Books.
Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation: Anxiety and
anger. New York: Basic Books.
Bowlby, J. (1980). Attachment and loss: Vol. 3. Loss: Sadness and depression. New York: Basic Books.
Breakwell, G. (1992). Social psychology of identity and the self concept.
San Diego, CA: Academic Press.
Brennan, K. A., & Morris, K. A. (1997). Attachment styles, self-esteem,
and patterns of seeking feedback from romantic partners. Personality
and Social Psychology Bulletin, 23, 23–31.
Bretherton, I. (1985). Attachment theory: Retrospect and prospect. Monographs for the Society for Research in Child Development, 50, 3–35.
Brewer, M. B. (1988). A dual process model of impression formation. In
T. K. Srull & R. S. Wyer Jr. (Eds.), A dual process model of impression
formation (pp. 1–36). Hillsdale, NJ: Erlbaum.
Brewer, M. B. (1991). The social self: On being the same and different at
the same time. Personality and Social Psychology Bulletin, 17, 475–
482.
Brewer, M. B., & Gardner, W. (1996). Who is this “we”? Levels of
collective identity and self representations. Journal of Personality and
Social Psychology, 71, 83–93.
Brewer, M. B., & Roccas, S. (2001). Individual values, social identity, and
optimal distinctiveness. In C. Sedikides & M. B. Brewer (Eds.), Individual self, relational self, collective self (pp. 219 –237). Philadelphia:
Taylor & Francis.
Brewer, M. B., & Weber, J. G. (1994). Self-evaluation effects of interpersonal versus intergroup social comparison. Journal of Personality and
Social Psychology, 66, 268 –275.
Bruner, J. (1990). Acts of meaning. Cambridge, MA: Harvard University
Press.
Bruner, J. S. (1957). Going beyond the information given. In H. E. Gruber,
K. R. Hammond, & R. Jessor (Eds.), Contemporary approaches to
cognition (pp. 41– 60). Cambridge, MA: Harvard University Press.
Bugental, D. B. (1992). Affective and cognitive processes within threatoriented family systems. In I. E. Sigel, A. McGillicuddy-de Lissi, & J.
Goodnow (Eds.), Parental belief systems: The psychological consequences for children (2nd ed., pp. 219 –248). Hillsdale, NJ: Erlbaum.
Bugental, D. B. (2000). Acquisition of algorithms of social life: A domainbased approach. Psychological Bulletin, 126, 187–219.
Cantor, N., & Kihlstrom, J. F. (1987). Personality and social intelligence.
Englewood Cliffs, NJ: Prentice Hall.
Cantor, N., Markus, H., Niedenthal, P., & Nurius, P. (1986). On motivation
and self-concept. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook
of motivation and cognition: Foundations of social behavior (pp. 96 –
121). New York: Guilford Press.
Cantor, N., & Mischel, W. (1979). Prototypicality and personality: Effects
on free recall and personality impressions. Journal of Research in
Personality, 13, 187–205.
Cantor, N., & Zirkel, S. (1990). Personality, cognition, and purposive
behavior. In L. A. Pervin (Ed.), Handbook of personality: Theory and
research (pp. 135–164). New York: Guilford Press.
Carlson, R. (1981). Studies in script theory: I. Adult analogues of a
childhood nuclear scene. Journal of Personality and Social Psychology, 40, 501–510.
Carson, R. C. (1969). Interaction concepts of personality. Chicago: Aldine
Publishing.
Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful conceptual
framework for personality-social, clinical, and health psychology. Psychological Bulletin, 92, 111–135.
Carver, C. S., & Scheier, M. F. (1998). On the self-regulation of behavior.
New York: Cambridge University Press.
Chen, S. (2001). The role of theories in mental representations and their use
in social perception: A theory-based approach to significant-other representations and transference. In G. B. Moskowitz (Ed.), Cognitive
social psychology: The Princeton Symposium on the Legacy and Future
of Social Cognition (pp. 125–142). Mahwah, NJ: Erlbaum.
Chen, S., & Andersen, S. M. (1999). Relationships from the past in the
present: Significant-other representations and transference in interpersonal life. In M. P. Zanna (Ed.), Advances in experimental social
psychology (Vol. 31, pp. 123–190). San Diego, CA: Academic Press.
Chen, S., Andersen, S. M., & Hinkley, K. (1999). Triggering transference:
Examining the role of applicability and use of significant-other representations in social perception. Social Cognition, 17, 332–365.
Chen, Y. R., Brockner, J., & Katz, T. (1998). Toward an explanation of
cultural differences in in-group favoritism: The role of individual versus
collective primacy. Journal of Personality and Social Psychology, 75,
1490 –1502.
Clark, M. S., & Reis, H. (1988) Interpersonal process in close relationships.
Annual Review of Psychology, 39, 609 – 672.
Clark, M. S., & Taraban, C. B. (1991) Reactions to and willingness to
express emotion in two types of relationships. Journal of Experimental
Social Psychology, 27, 324 –336.
Collins, N. L., & Read, S. J. (1994). Cognitive representations of attachment: The structure and function of working models. In K. Bartholomew
(Ed.), Attachment processes in adulthood (pp. 53–90). Bristol, PA:
Jessica Kingsley.
Conway, M., & Ross, M. (1984). Getting what you want by revising what
you had. Journal of Personality and Social Psychology, 47, 738 –748.
Cooley, C. H. (1902). Human nature and social order. New York: Scribner.
Crocker, J., & Luhtanen, R. (1990). Collective self-esteem and ingroup
bias. Journal of Personality and Social Psychology, 58, 60 – 67.
Crocker, J., Luhtanen, R., Blaine, B., & Broadnax, S. (1994). Collective
self-esteem and psychological well-being among White, Black, and
Asian college students. Personality and Social Psychology Bulletin, 20,
503–513.
THE RELATIONAL SELF
Crocker, J., & Wolfe, C. (2001). Contingencies of self-worth. Psychological Review, 108, 593– 623.
Cross, S. E., & Madson, L. (1997). Models of the self: Self-construal and
gender. Psychological Bulletin, 122, 5–37.
Deaux, K. (1992). Focusing on the self: Challenges to self-definition and
their consequences for mental health. In D. N. Ruble & P. R. Costanzo
(Eds.), The social psychology of mental health: Basic mechanisms and
applications (pp. 301–327). New York: Guilford Press.
Deaux, K. (1993). Reconstructing social identity. Personality and Social
Psychology Bulletin, 19, 4 –12.
Deci, E. L. (1995). Why we do what we do. New York: Putnam.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and selfdetermination in human behavior. New York: Plenum Press.
Deci E. L., & Ryan, R. M. (1991). A motivational approach to self:
Integration in personality. In R. Dienstbier (Ed.), Nebraska Symposium
on Motivation: Vol. 38. Perspectives on motivation (pp. 237–288).
Lincoln: University of Nebraska Press.
Deci E. L., & Ryan, R. M. (2000). The “what” and “why” of goal pursuits:
Human needs and the self-determination of behavior. Psychological
Inquiry, 11, 227–268.
Demo, D. H. (1992). The self-concept over time: Research issues and
directions. Annual Review of Sociology, 18, 303–326.
Dodge, K. & Price, J. M. (1994). On the relation between social information processing and socially competent behavior in early school-aged
children. Child Development, 65, 1385–1397.
Downey, G., & Feldman, S. (1996). Implications of rejection sensitivity for
intimate relationships. Journal of Personality and Social Psychology, 70, 1327–1343.
Dweck, C. S., & Leggett, E. L. (1988). A social– cognitive approach to
motivation and personality. Psychological Review, 95, 256 –273.
Ehrenreich, J. H. (1989). Transference: One concept or many? Psychoanalytic Review, 76, 37– 65.
Endler, N. S. (1984). Interactionism. In N. S. Endler & J. M. Hunt (Eds.),
Personality and the behavioral disorders (Vol. 1). New York: Wiley.
Epstein, S. (1973). The self-concept revisited or a theory of a theory.
American Psychologist, 28, 405– 416.
Epstein, S. (1983). Aggregation and beyond: Some basic issues on the
prediction of behavior. Journal of Personality, 51, 360 –392.
Erber, R., & Erber, M. W. (2000). The self-regulation of moods: Second
thoughts on the importance of happiness in everyday life. Psychological
Inquiry, 51, 360 –392.
Fairbairn, W. R. D. (1952). Psychoanalytic studies of personality. London:
Tavistock.
Fiske, A. P. (1992). The four elementary forms of sociality: Framework for
a unified theory of social relations. Psychological Review, 99, 689 –723.
Fiske, S. T., & Pavelchak, M. (1986). Category-based versus piecemealbased affective responses: Developments in schema-triggered affect. In
R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of motivation and
cognition (pp. 167–203). New York: Guilford Press.
Fiske, S. T., & Taylor, S. E. (1991). Social cognition (2nd ed.). New York:
McGraw-Hill.
Frankl, V. E. (1959). Man’s search for meaning. Boston, MA: Beacon
Press.
Freud, S. (1958). The dynamics of transference. In J. Strachey (Ed. &
Trans.), The standard edition of the complete psychological works of
Sigmund Freud (Vol. 12, pp. 97–108). London: Hogarth. (Original work
published 1912)
Gaertner, L., Sedikides, C., & Graetz, K. (1999). In search of self-definition: Motivational primacy of the individual self, motivational primacy
of the collective self, or other contextual primacy? Journal of Personality and Social Psychology, 76, 5–18.
Gergen, K. J., & Gergen, M. M. (1988). Narrative and the self as relationship. In L. Berkowitz (Ed.), Advances in experimental social psychol-
641
ogy: Vol. 21. Social psychological studies of the self: Perspectives and
programs (pp. 17–56). San Diego, CA: Academic Press.
Glassman, N. S., & Andersen, S. M. (1999a). Activating transference
without consciousness: Using significant-other representations to go
beyond what is subliminally given. Journal of Personality and Social
Psychology, 77, 1146 –1162.
Glassman, N. S., & Andersen, S. M. (1999b). Streams of thought about the
self and significant others: Transference as the construction of interpersonal meaning. In J. A. Singer & P. Salovey (Eds.), At play in the fields
of consciousness (pp. 103–140). Mahwah, NJ: Erlbaum.
Glassman, N. S., & Andersen, S. M. (1999c). Transference in social
cognition: Persistence and exacerbation of significant-other-based inferences over time. Cognitive Therapy and Research, 23, 75–91.
Glendin, E. T. (1962). Experiencing and the creation of meaning. Glencoe,
IL: Free Press.
Greenberg, J., & Pyszczynski, T. (1985). Compensatory self-inflation: A
response to the threat to self-regard of public failure. Journal of Personality and Social Psychology, 49, 273–280.
Greenberg, J. R., & Mitchell, S. A. (1983). Object relations in psychoanalytic theory. Cambridge, MA: Harvard University Press.
Greenwald, A. G. (1980). The totalitarian ego. American Psychologist, 35,
603– 618.
Greenwald, A. G., & Banaji, M. (1989). The self as a memory system:
Powerful, but ordinary. Journal of Personality and Social Psychology, 57, 41–54.
Greenwald, A. G., & Pratkanis, A. R. (1984). The self. In R. S. Wyer &
T. K. Srull (Eds.), Handbook of social cognition (Vol. 3, pp. 129 –178).
Hillsdale, NJ: Erlbaum.
Griffin, D., & Bartholomew, K. (1994). Models of self and other: Fundamental dimensions underlying measures of adult attachment. Journal of
Personality and Social Psychology, 67, 430 – 445.
Guidano, V. F., & Liotti, G. (1983). Cognitive processes and emotional
disorders. New York: Guilford.
Guisinger, S., & Blatt, S. J. (1994). Individuality and relatedness: Evolution of a fundamental dialectic. American Psychologist, 49, 104 –111.
Hardin, C. D. & Higgins, E. T. (1996). Shared reality: How social verification makes the subjective objective. In R.M. Sorrentino & E.T. Higgins (Eds.), Handbook of motivation and cognition (Vol. 3, pp. 28 – 84).
New York: Guilford Press.
Hardin, C. D., & Rothman, A. J. (1997). Rendering accessible information
relevant: The applicability of everyday life. In R. J. Wyer Jr. (Ed.),
Advances in social cognition (Vol. 10, pp. 143–156). Mahwah, NJ:
Erlbaum.
Harter, S. (1999). The construction of the self: A developmental perspective. New York: Guilford Press.
Harter, S., Waters, P., & Whitesell, N. R. (1998). Relational self-worth:
Differences in perceived worth as a person across interpersonal contexts
among adolescents. Child Development, 69, 756 –766.
Haslam, N. (1994). Mental representation of social relationships: Dimensions, laws, or categories? Journal of Personality and Social Psychology, 67, 575–584.
Hazan, C., & Shaver, P. (1994). Attachment as an organizational framework for research on close relationships. Psychological Inquiry, 5, 1–22.
Heidegger, M. (1962). Being and time. New York: Harper & Row.
Helgeson, V. S. (1994). Relation of agency and communion to well-being:
Evidence and potential explanations. Psychological Review, 116, 412–
428.
Higgins, E. T. (1987). Self-discrepancy theory: A theory relating self and
affect. Psychological Review, 94, 319 –340.
Higgins, E. T. (1989a). Continuities and discontinuities in self-regulatory
and self-evaluative processes: A developmental theory relating self and
affect. Journal of Personality, 57, 407– 444.
Higgins, E. T. (1989b). Knowledge accessibility and activation: Subjectivity and suffering from unconscious sources. In J. S. Uleman & J. A.
642
ANDERSEN AND CHEN
Bargh (Eds.), Unintended thought (pp. 75–123). New York: Guilford
Press.
Higgins, E. T. (1990). Personality, social psychology, and person–situation
relations: Standards and knowledge activation as a common language. In
L. A. Pervin (Ed.), Handbook of personality (pp. 301–338). New York:
Guilford Press.
Higgins, E. T. (1991). Development of self-regulatory and self-evaluative
processes: Costs, benefits, and tradeoffs. In M. R. Gunnar & L. A.
Stroufe (Eds.), Self processes and development: The Minnesota Symposia on Child Development (Vol. 23, pp. 125–165). Hillsdale, NJ: Erlbaum.
Higgins, E. T. (1996a). Emotional experiences: The pains and pleasures of
distinct regulatory systems. In R. D. Kavanaugh, B. Zimmerberg, & S.
Fein (Eds.), Emotion: Interdisciplinary perspectives (pp. 203–241).
Mahwah, NJ: Erlbaum.
Higgins, E. T. (1996b). Ideals, oughts, and regulatory focus: Affect and
motivation from distinct pains and pleasures. In P. M. Gollwitzer & J. A.
Bargh (Eds.), The psychology of action: Linking cognition and motivation to behavior (pp. 91–114). New York: Guilford Press.
Higgins, E. T. (1996c). Knowledge activation: Accessibility, applicability,
and salience. In E. T. Higgins & A. W. Kruglanski (Eds.), Social
psychology: Handbook of basic principles (pp. 133–168). New York:
Guilford Press.
Higgins, E. T. (1996d). The “self-digest”: Self-knowledge serving selfregulatory functions. Journal of Personality and Social Psychology, 71,
1062–1083.
Higgins, E. T., & Bargh, J. A. (1987). Social cognition and social perception. In M. R. Rosenzweig & L. W. Porter (Eds.), Annual Review of
Psychology (Vol. 38, pp. 369 – 425). Palo Alto, CA: Annual Reviews.
Higgins, E. T., Bond, R. N., Klein, R., & Strauman, T. (1986). Selfdiscrepancies and emotional vulnerability: How magnitude, accessibility, and type of discrepancy influence affect. Journal of Personality and
Social Psychology, 51, 5–15.
Higgins, E. T., & Brendl, C. M. (1995). Accessibility and applicability:
Some “activation rules” influencing judgment. Journal of Experimental
Social Psychology, 31, 218 –243.
Higgins, E. T., & King, G. A. (1981). Accessibility of social constructs:
Information-processing consequences of individual and contextual variability. In N. Cantor & J. F. Kihlstrom (Eds.), Personality, cognition,
and social interaction (pp. 69 –121). Hillsdale, NJ: Erlbaum.
Higgins, E. T., King, G. A., & Mavin, G. H. (1982). Individual construct
accessibility and subjective impressions and recall. Journal of Personality and Social Psychology, 43, 35– 47.
Higgins, E. T., Loeb, I., & Moretti, M. (1995). Self-discrepancies and
developmental shifts in vulnerability: Life transitions in the regulatory
significance of others. In D. Cicchetti & S. L. Toth (Eds.), Emotion,
cognition, and representation (pp. 191–230). Rochester, NY: University
of Rochester Press.
Higgins, E. T., & May, D. (2001). Individual self-regulatory functions: It’s
not “we” regulation, but it’s still social. In C. Sedikides & M. B. Brewer
(Eds.), Individual self, relational self, collective self (pp. 47– 67). Philadelphia: Taylor & Francis.
Higgins, E. T., Rholes, W. S., & Jones, C. R. (1977). Category accessibility
and impression formation. Journal of Experimental Social Psychology, 13, 141–154.
Higgins, E. T., Roney, C. J., Crowe, E., & Hymes, C. (1994). Ideal versus
ought predictions for approach and avoidance: Distinct self-regulatory
systems. Journal of Personality and Social Psychology, 66, 276 –286.
Higgins, E. T., Van Hook, E., & Dorfman, D. (1988). Do self-attributes
form a cognitive structure? Social Cognition, 6, 177–207.
Hinkley, K., & Andersen, S. M. (1996). The working self-concept in
transference: Significant-other activation and self-change. Journal of
Personality and Social Psychology, 71, 1279 –1295.
Holmes, J. G., & Rempel, J. K. (1989). Trust in close relationships. Review
of Personality and Social Psychology, 10, 187–219.
Horney, K. (1939). New ways in psychoanalysis. New York: Norton.
Horney, K. (1945). Our inner conflicts. New York: Norton.
James, W. (1890). The principles of psychology (Vol. 1). Cambridge, MA:
Harvard University Press.
Janoff-Bulman, R. (1992). Shattered assumptions: Towards a new psychology of trauma. New York: Free Press.
Johnson, J. T., & Boyd, K. R. (1995). Dispositional traits versus the content
of experience: Actor/observer differences in judgments of the “authentic
self. ” Personality and Social Psychology Bulletin, 21, 375–383.
Jung, C. G. (1933). Modern man in search of a soul. New York: Harcourt,
Brace.
Karylowski, J. J., Konarzewski, K., & Motes, M. (2000). Recruitment of
exemplars as reference points in social judgments. Journal of Experimental Social Psychology, 36, 275–303.
Kelly, G. A. (1955). The psychology of personal constructs. New York:
Norton.
Kitayama, S., Markus, H. R., Matsumoto, H., & Norasakkunkit, V. (1997).
Individual and collective processes in the construction of self: Selfenhancement in the United States and self-criticism in Japan. Journal of
Personality and Social Psychology, 72, 1245–1267.
Klein, M. (1932). The psychoanalysis of children. London: Hogarth Press.
Klinger, E. (1977). Meaning and void. Minneapolis: University of Minnesota Press.
Kruglanski, A. W. (1996). Motivated social cognition: Principles of the
interface. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: A handbook of basic principles (pp. 493–522). New York: Guilford
Press.
Lamiell, J. T. (1981). Toward an idiothetic psychology of personality.
American Psychologist, 36, 276 –289.
Larsen, R. J. (2000). Toward a science of mood regulation. Psychological
Inquiry, 11, 129 –141.
Leary, M. R., Tambor, E. S., Terdal, S. K., & Downs, D. L. (1995).
Self-esteem as an interpersonal monitor: The sociometer hypothesis.
Journal of Personality and Social Psychology, 68, 518 –530.
Levine, S. R., Wyer, R. S., & Schwartz, N. (1994). Are you what you feel?
The affective and cognitive determinants of self-judgments. European
Journal of Social Psychology, 24, 63–77.
Linville, P. W. (1985). Self-complexity and affective extremity: Don’t put
all your eggs in one cognitive basket. Social Cognition, 3, 94 –120.
Linville, P. W., & Carlston, D. E. (1994). Social cognition of the self. In
P. G. Devine, D. C. Hamilton, & T. M. Ostrom (Eds.), Social cognition:
Impact on social psychology (pp. 143–193). New York: Academic Press.
Linville, P. W., & Fischer, G. W. (1993). Exemplar and abstraction models
of perceived group variability and stereotypicality. Social Cognition, 11,
92–125.
Loftus, E. F. (1982). Memory and its distortions. In A. G. Kraut (Ed.), The
G. Stanley Hall lecture series (Vol. 2, pp. 123–154). Washington, DC:
American Psychological Association.
Loftus, E. F., & Greene, E. (1980). Warning: Even memory for faces may
be contagious. Law and Human Behavior, 4, 323–334.
Luborsky, L., & Crits-Cristoph, P. (1990). Understanding transference:
The CCRT method. New York: Basic Books.
Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the
family: Parent– child interaction. In P. H. Mussen (Ed.), Handbook of
child psychology: Vol. IV. Socialization, personality, and social development (pp. 1–101). New York: Wiley.
Magnussen, D. (1990). Personality development from an interactional
perspective. In L. A. Pervin (Ed.), Handbook of personality (pp. 193–
222). New York: Guilford Press.
Markman, A. B., & Gentner, D. (in press). Nonintentional similarity
processing. In R. Hassan, J. A. Bargh, & J. S. Uleman (Eds.), The new
unconscious. Oxford, England: Oxford University Press.
THE RELATIONAL SELF
Markus, H. (1977). Self-schemata and processing information about the
self. Journal of Personality and Social Psychology, 35, 63–78.
Markus, H., & Cross, S. (1990). The interpersonal self. In L. A. Pervin,
(Ed.), Handbook of personality (pp. 576 – 608). New York: Guilford
Press.
Markus, H., & Kitayama, S. (1991). Culture and the self: Implications for
cognition, emotion, and motivation. Psychological Review, 98, 224 –
253.
Markus, H., & Nurius, P. (1986). Possible selves. American Psychologist, 41, 954 –969.
Markus, H., & Wurf, E. (1987). The dynamic self-concept: A social–
psychological perspective. Annual Review of Psychology, 38, 299 –337.
Martin, L. L., Tesser, A., & McIntosh, W. D. (1993). Wanting but not
having: The effects of unattained goals on thoughts and feelings. In
D. M. Wegner & J. W. Pennebaker (Eds.), The handbook of mental
control (pp. 552–572). New York: Prentice Hall.
McAdams, D. P. (1980). A thematic coding system for the intimacy
motive. Journal of Research in Personality, 14, 413– 432.
McAdams, D. P. (1985). Power, intimacy, and the life story: Personological inquiries into identity. New York: Guilford Press.
McAdams, D. P. (1989). Intimacy: The need to be close. New York:
Doubleday.
McGuire, W. J., McGuire, C. V., & Cheever, J. (1986). The self in society:
Effects of social contexts on the sense of self. British Journal of Social
Psychology, 25, 259 –270.
Mead, G. H. (1934). Mind, self, and society: From the standpoint of a
social behaviorist. Chicago: University of Chicago Press.
Mikulincer, M. (1995). Attachment style and the mental representation of
the self. Journal of Personality and Social Psychology, 69, 1203–1215.
Mikulincer, M. (1998). Adult attachment style and affect regulation: Strategic variations in self-appraisals. Journal of Personality and Social
Psychology, 75, 420 – 435.
Mills, J., & Clark, M. S. (1994). Communal and exchange relationships:
Controversies and research. In R. Erber & R. Gilmour (Eds.), Theoretical frameworks for personal relationships (pp. 29 – 42). Hillsdale, NJ:
Erlbaum.
Mischel, W. (1968). Personality and assessment. New York: Wiley.
Mischel, W. (1973). Toward a cognitive social learning reconceptualization of personality. Psychological Review, 80, 252–253.
Mischel, W. (1990). Personality dispositions revisited and revised: A view
after three decades. In L. A. Pervin (Ed.), Handbook of personality:
Theory and research (pp. 111–134). New York: Guilford Press.
Mischel, W. (1999). Personality coherence and dispositions in a cognitive–
affective personality (CAPS) approach. In D. Cervone & Y. Shoda
(Eds.), The coherence of personality: Social– cognitive bases of consistency, variability, and organization (pp. 37– 60). New York: Guilford
Press.
Mischel, W., Cantor, N., & Feldman, S. (1996). Principles of self-regulation: The nature of willpower and self-control. In E. T. Higgins & A. W.
Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp.
329 –360). New York: Guilford Press.
Mischel, W., & Shoda, Y. (1995). A cognitive-affective system theory of
personality: Reconceptualizing situations, dispositions, dynamics, and
invariance in personality structure. Psychological Review, 102, 246 –
268.
Moretti, M. M., & Higgins, E. T. (1990). Relating self-discrepancy to
self-esteem: The contribution of discrepancy beyond actual self-ratings.
Journal of Experimental Social Psychology, 26, 108 –123.
Moretti, M. M., & Higgins, E. T. (1999). Own versus other standpoints in
self-regulation: Developmental antecedents and functional consequences. Review of General Psychology, 3, 188 –223.
Murray, S. L., & Holmes, J. G. (1993). Seeing virtues in faults: Negativity
and the transformation of interpersonal narratives in close relationships.
Journal of Personality and Social Psychology, 65, 707–722.
643
Murray, S. L., & Holmes, J. G. (1994). Storytelling in close relationships:
The construction of confidence. Personality and Social Psychology
Bulletin, 20, 650 – 663.
Niedenthal, P. M., & Beike, D. R. (1997). Interrelated and isolated selfconcepts. Personality & Social Psychology Review, 1, 106 –128.
Nisbett, R. E., & Ross, L. (1980). Human inference: Strategies and
shortcomings of social judgment. Englewood Cliffs, NJ: Prentice Hall.
Oatley, K., & Bolton, W. (1985). A social– cognitive theory of depression
in reaction to life events. Psychological Review, 92, 372–388.
Ogilvie, D. M., & Ashmore, R. D. (1991). Self-with-other representation as
a unit of analysis in self-concept research. In R. C. Curtis (Ed.), The
relational self: Theoretical convergencies in psychoanalysis and social
psychology (pp. 282–314). New York: Guilford Press.
Park, C. L., & Folkman, S. (1997). Meaning and the context of stress and
coping. Review of General Psychology, 1, 115–144.
Pelham, B. W. (1993). The idiographic nature of human personality:
Examples of the idiographic self-concept. Journal of Personality and
Social Psychology, 64, 665– 677.
Pennebaker, J. W. (in press). Writing about emotional experiences as a
therapeutic process. Psychological Science.
Pervin, L. A. (1985). Personality: Current controversies, issues, and directions. Annual Review of Psychology, 36, 83–114.
Pervin, L. A., & Lewis, M. (Eds.). (1978). Perspectives in interactional
psychology. New York: Plenum Press.
Prentice, D. (1990). Familiarity and differences in self-and-other representations. Journal of Personality and Social Psychology, 59, 369 –383.
Read, S. J., Vanman, E. J., & Miller, L. C. (1997). Connectionism, parallel
constraint satisfaction processes, and Gestalt principles: (Re)introducing
cognitive dynamics to social psychology. Personality and Social Psychology Review, 1, 26 –53.
Reid, A., & Deaux, K. (1996). Relationship between social and personal
identities: Segregation or integration. Journal of Personality and Social
Psychology, 71, 1084 –1091.
Reznik, I., & Andersen, S. M. (2001). Self-discrepancy theory and
significant-other representations: Eliciting specific negative affects by
triggering everyday transference. Unpublished manuscript, New York
University.
Rhee, E., Uleman, J. S., Lee, H. K., & Roman, R. J. (1995). Spontaneous
self-descriptions and ethnic identities in individualistic and collectivistic
cultures. Journal of Personality and Social Psychology, 69, 142–152.
Rogers, C. (1951). Client-centered therapy. Boston: Houghton-Mifflin.
Rosenberg, M. (1979). Conceiving the self. New York: Basic Books.
Rosenthal, R. (1994). Interpersonal expectancy effects: A 30-year perspective. Current Directions in Psychological Science, 3, 176 –179.
Safran, J. D. (1990). Toward a refinement of cognitive therapy in light of
interpersonal theory: I. Theory. Clinical Psychology Review, 10, 87–105.
Sande, G. N., Goethals, G. R., & Radloff, C. E. (1988). Perceiving one’s
own traits and others’: The multifaceted self. Journal of Personality and
Social Psychology, 54, 13–20.
Sarbin, T. R., & Allen, V. L. (1968). Role theory. In G. Lindzey & E.
Aronson (Eds.), The handbook of social psychology (Vol. 1, pp. 488 –
567). Reading, MA: Addison Wesley.
Schank, R. C., & Abelson, R. P. (1977). Scripts, plans, goals, and understanding: An inquiry into human knowledge structures. Hillsdale, NJ:
Erlbaum.
Schwarz, N. (1990). Feelings as information: Informational and motivational functions of affective states. In E. T. Higgins & R. M. Sorrentino
(Eds.), Handbook of motivation and cognition: Foundations of social
behavior (Vol. 2, pp. 527–561). New York: Guilford Press.
Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments
of well-being: Informative and directive functions of affective states.
Journal of Personality and Social Psychology, 45, 513–523.
Sedikides, C., & Gaertner, L. (2001). The social self: The quest for
self-definition and the motivational primacy of the individual self. In
644
ANDERSEN AND CHEN
J. P. Forgas, K. D. Williams, & L. Wheeler (Eds.), The social mind:
Cognitive and motivation aspects of interpersonal behavior (pp. 115–
138). New York: Cambridge University Press.
Sedikides, C., Olsen, N., & Reis, H. T. (1993). Relationships as natural
categories. Journal of Personality and Social Psychology, 64, 71– 82.
Sedikides, C., & Skowronski, J. J. (1990). Toward reconciling personality
and social psychology; A construct accessibility approach. Journal of
Social Behavior and Personality, 5, 531–546.
Sedikides, C., & Skowronski, J. J. (1991). The law of cognitive structure
activation. Psychological Inquiry, 2, 169 –184.
Seligman, M. E. P. (1975). Helplessness: On depression, development, and
death. San Francisco: Freeman.
Seta, J. J., & Seta, C. E. (1996). Big fish in small ponds: A social hierarchy
analysis of intergroup bias. Journal of Personality and Social Psychology, 71, 1210 –1221.
Shoda, Y., Mischel, W., & Wright, J. C. (1989). Intuitive interactionism in
person perception: Effects of situation– behavior relations on dispositional judgments. Journal of Personality and Social Psychology, 56,
41–53.
Shoda, Y., Mischel, W., & Wright, J. C. (1994). Intraindividual stability in
the organization and patterning of behavior: Incorporating psychological
situations into the idiographic analysis of personality. Journal of Personality and Social Psychology, 67, 674 – 687.
Showers, C. (1992). Compartmentalization of positive and negative selfknowledge: Keeping bad apples out of the bunch. Journal of Personality
and Social Psychology, 62, 1036 –1049.
Showers, C. J., & Kevlyn, S. B. (1999). Organization of knowledge about
a relationship partner: Implications for liking and loving. Journal of
Personality and Social Psychology, 76, 958 –971.
Silver, R. C., & Wortman, C. B. (1980). Coping with undesirable life
events. In J. Garber & M. E. P. Seligman (Eds.), Human helplessness
(pp. 279 –340). New York: Academic Press.
Simon, B., Pantaleo, G., & Mummendey, A. (1995). Unique individual or
interchangeable group member? The accentuation of intragroup differences versus similarities as an indicator of the individual self versus the
collective self. Journal of Personality and Social Psychology, 69, 106 –
119.
Simpson, J. A., & Rholes, W. S. (1998). Attachment theory and close
relationships. New York: Guilford Press.
Singer, J. L. (1988). Reinterpreting the transference. In D. C. Turk & P.
Salvey (Eds.), Reasoning, interference, and judgment in clinical psychology (pp. 182–205). New York: Free Press.
Smith, E. E., Patalino, A. L., & Jonides, J. (1998). Alternative strategies of
categorization. Cognition, 65, 167–196.
Smith, E. R. (1990). Content and process specificity in the effects of prior
experiences. In T. K. Srull & R. S. Wyer Jr. (Eds.), Content and process
specificity in the effects of prior experiences (pp. 1–59). Hillsdale, NJ:
Erlbaum.
Smith, E. R. (1992). The role of exemplars in social judgment. In L. L.
Martin & A. Tesser (Eds.), The construction of social judgments (pp.
107–132). Hillsdale, NJ: Erlbaum.
Smith, E. R. (1996). What do connectionism and psychology have to offer
each other? Journal of Personality and Social Psychology, 70, 893–912.
Smith, E. R. (1998). Mental representation and memory. In D.T. Gilbert &
S.T. Fiske (Eds.), The handbook of social psychology (Vol. 1, 4th ed.,
pp. 391– 445). New York: McGraw-Hill.
Smith, E. R., & Henry, S. (1996). An in-group becomes part of the self:
Response time evidence. Personality and Social Psychology Bulletin, 22, 635– 642.
Smith, E. R., Murphy, J., & Coats, S. (1999). Attachment to groups:
Theory and management. Journal of Personality and Social Psychology, 77, 94 –110.
Smith, E. R., Stewart, T. L., & Buttram, R. T. (1992). Inferring a trait from
a behavior has long-term, specific effects. Journal of Personality and
Social Psychology, 62, 753–759.
Smith, E. R., & Zarate, M. A. (1990). Exemplar and prototype use in social
categorization. Social Cognition, 8, 243–262.
Smith, E. R., & Zarate, M. A. (1992). Exemplar-based model of social
judgment. Psychological Review, 99, 3–21.
Snyder, M. (1992). Motivational foundations of behavioral confirmation.
In M. P. Zanna (Ed.), Advances in experimental social psychology
(Vol. 25, pp. 67–114). New York: Academic Press.
Snyder, M., & Cantor, N. (1998). Understanding personality and social
behavior: A functionalist strategy. In D. T. Gilbert & S. T. Fiske (Eds.),
The handbook of social psychology (Vol. 1, 4th ed., pp. 635– 679). New
York: McGraw-Hill.
Snyder, M., Tanke, E. D., & Berscheid, E. (1977). Social perception and
interpersonal behavior: On the self-fulfilling nature of social stereotypes.
Journal of Personality and Social Psychology, 35, 656 – 666.
Srull, T. K., & Wyer, R. S. (1979). The role of category accessibility in the
interpretation of information about persons: Some determinants and
implications. Journal of Personality and Social Psychology, 37, 1660 –
1672.
Steele, C. M. (1988). The psychology of self-affirmation: Sustaining the
integrity of the self. In L. Berkowitz (Ed.), Advances in experimental
social psychology (Vol. 21, pp. 261–302). New York: Academic Press.
Strauman, T. J. (1996). Stability within the self: A longitudinal study of the
structural implication of self-discrepancy theory. Journal of Personality
and Social Psychology, 71, 1142–1153.
Strauman, T., & Higgins, E. T. (1987). Automatic activation of selfdiscrepancies and emotional syndromes: When cognitive structures influence affect. Journal of Personality and Social Psychology, 53, 1004 –
1014.
Strauman, T., & Higgins, E. T. (1988). Self-discrepancies as predictors of
vulnerability to distinct syndromes of chronic emotional distress. Journal of Personality, 56, 685–707.
Sullivan, H. S. (1940). Conceptions in modern psychiatry. New York:
Norton.
Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York:
Norton.
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup
conflict. In W. G. Austin & S. Worchel (Eds.), The social psychology of
intergroup relations (pp. 33– 47). Monterey, CA: Brooks/Cole.
Taylor, S. E. (1991). Asymmetrical effects of positive and negative events:
The mobilization–minimization hypothesis. Psychological Bulletin, 110,
67– 85.
Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social
psychological perspective on mental health. Psychological Bulletin, 103,
193–210.
Taylor, S. E., & Lobel, M. (1989). Social comparison activity under threat:
Downward evaluation and upward contrasts. Psychological Review, 96,
569 –575.
Tesser, A. (1988). Toward a self-evaluation maintenance model of social
behavior. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 21, pp. 181–227). New York: Academic Press.
Tesser, A., Millar, M., & Moore, J. (1988). Some affective consequences
of social comparison and reflection processes: The pain and pleasure of
being close. Journal of Personality and Social Psychology, 54, 49 – 61.
Thompson, R. A. (1998). Early sociopersonality development. In W.
Damon (Series Ed.) & N. Eisenberg (Vol. Ed.), Handbook of child
psychology: Vol. 3. Social, emotional, and personality development (5th
ed., pp. 25–104). New York: Wiley.
Tice, D. M., & Bratslavsky, E. (2000). Giving in to feel good: The place
of emotion regulation in the context of general self-control. Psychological Inquiry, 11, 149 –159.
Tomkins, S. S. (1979). Script theory: Differential magnification of affects.
In H. E. Howe Jr. & R. A. Dienstbier (Eds.), Nebraska Symposium on
THE RELATIONAL SELF
Motivation (Vol. 26, pp. 201–236). Lincoln: University of Nebraska
Press.
Trafimow, D., Silverman, E. S., Fan, R. M., & Law, J. S. F. (1997). The
effects of language and priming on the relative accessibility of the
private self and the collective self. Journal of Cross-Cultural Psychology, 28, 107–123.
Trafimow, D., Triandis, H. C., & Goto, S. (1991). Some tests of the
distinction between the private self and the collective self. Journal of
Personality and Social Psychology, 60, 649 – 655.
Triandis, H. C. (1988). Cross-cultural contributions to theory in social
psychology. In M. H. Bond (Ed.), The cross-cultural challenge to social
psychology (pp. 122–140). Thousand Oaks, CA: Sage.
Triandis, H. C. (1989). The self and social behavior in differing cultural
contexts. Psychological Review, 96, 506 –520.
Turner, J. C., Oakes, P. J., Haslam, S. A., & McGarty, C. (1994). Self and
collective: Cognition and social context. Personality and Social Psychology Bulletin, 20, 454 – 463.
Wachtel, P. L. (1981). Transference, schema, and assimilation: The relevance of Piaget to the psychoanalytic theory of transference. In P. L.
Wachtel (Ed.), Action and insight (pp. 26 – 44). New York: Guilford
Press.
Wegner, D. (1995). A computer model of human transactive memory.
Social Cognition, 13, 319 –339.
Westen, D. (1988). Transference and information processing. Clinical
Psychology Review, 8, 161–179.
645
White, R. W. (1959). Motivation reconsidered: The concept of competence. Psychological Review, 66, 297–333.
Wright, J. C., & Mischel, W. (1987). A conditional approach to dispositional constructs: The local predictability of social behavior. Journal of
Personality and Social Psychology, 53, 1159 –1177.
Wright, J. C., & Mischel, W. (1988). Conditional hedges and the intuitive
psychology of traits. Journal of Personality and Social Psychology, 55,
454 – 469.
Wright, S. C., Aron, A., McLaughlin-Volpe, T., & Ropp, S. A. (1997). The
extended contact effect: Knowledge of cross-group friendships and
prejudice. Journal of Personality and Social Psychology, 73, 73–90.
Wyer, R. S., Clore, G. L., & Isbell, L. (1999). Affect and information
processing. In M. Zanna (Ed.), Advances in experimental social psychology (Vol. 31, pp. 1–77). New York: Academic Press.
Wyer, R. S., & Srull, T. K. (1986). Human cognition in its social context.
Psychological Review, 93, 322–359.
Wylie, R. C. (1961). The self-concept: A critical survey of pertinent
research literature. Lincoln: University of Nebraska Press.
Wylie, R. C. (1974). The self-concept: A review of methodological considerations and measuring instruments. Lincoln: University of Nebraska
Press.
Received October 29, 1999
Revision received September 10, 2001
Accepted September 11, 2001 䡲